Search results for: deep brain stimulation

Authors: Anum Ali, Kashaf ad Dooja, Asif Saleem

Abstract:

The future smart cities trend will be towards Internet of Things (IoT); IoT creates dynamic connections in a ubiquitous manner. Smart cities offer ease and flexibility for daily life matters. By using small devices that are connected to cloud servers based on IoT, network traffic between these devices is growing exponentially, whose security is a concerned issue, since ratio of cyber attack may make the network traffic vulnerable. This paper discusses the latest machine learning approaches in related work further to tackle the increasing rate of cyber attacks, machine learning algorithm is applied to IoT-based network traffic data. The proposed algorithm train itself on data and identify different sections of devices interaction by using supervised learning which is considered as a classifier related to a specific IoT device class. The simulation results clearly identify the attacks and produce fewer false detections.

Keywords: IoT, traffic security, deep learning, classification

Procedia PDF Downloads 137

Authors: Ubon Asuquo Essien, Okwudili Bismark Ibeagwa, Daberechi Peace Ubabuko

Abstract:

The Broiler Poultry subsector is dominated by small scale production with low aggregate output. The high cost of inputs currently experienced in Nigeria tends to aggravate the situation; hence many broiler farmers struggle to break-even. This study was designed to examine income and input factors in small scale deep liter broiler production in Imo state, Nigeria. Specifically, the study examined; socio-economic characteristics of small scale deep liter broiler producing Poultry farmers; estimate cost and returns of broiler production in the area; analyze input factors in broiler production in the area and examined marketability, age and profitability of the enterprise. A multi-stage sampling technique was adopted in selecting 60 small scale broiler farmers who use deep liter system from 6 communities through the use of structured questionnaire. The socioeconomic characteristics of the broiler farmers and the profitability/ marketability age of the birds were described using descriptive statistical tools such as frequencies, means and percentages. Gross margin analysis was used to analyze the cost and returns to broiler production, while Cobb Douglas production function was employed to analyze input factors in broiler production. The result of the study revealed that the cost of feed (P<0.1), deep liter material (P<0.05) and medication (P<0.05) had a significant positive relationship with the gross return of broiler farmers in the study area, while cost of labour, fuel and day old chicks were not significant. Furthermore, Gross profit margin of the farmers who market their broiler at the 8th week of rearing was 80.7%; and 78.7% and 60.8% for farmers who market at the 10th week and 12th week of rearing, respectively. The business is, therefore, profitable but at varying degree. Government and Development partners should make deliberate efforts to curb the current rise in the prices of poultry feeds, drugs and timber materials used as bedding so as to widen the profit margin and encourage more farmers to go into the business. The farmers equally need more technical assistance from extension agents with regards to timely and profitable marketing.

Keywords: broilers, factor analysis, income, small scale

Procedia PDF Downloads 61

Authors: K. Biswas, U. H. Armin, S. M. J. Prodhan, J. A. Prithul, S. Sarker, F. Afrin

Abstract:

Alzheimer’s disease (AD) (a progressive neurodegenerative disorder) is mostly predominant cause of dementia in the elderly. Prolonging the function of acetylcholine by inhibiting both acetylcholinesterase and butyrylcholinesterase is most effective treatment therapy of AD. Traditionally Pterocarpus santalinus L. is widely known for its medicinal use. In this study, in vitro acetylcholinesterase inhibitory activity was investigated and methanolic extract of the plant showed significant activity. To confirm this activity (in vivo), learning and memory enhancing effects were tested in mice. For the test, memory impairment was induced by scopolamine (cholinergic muscarinic receptor antagonist). Anti-amnesic effect of the extract was investigated by the passive avoidance task in mice. The study also includes brain acetylcholinesterase activity. Results proved that scopolamine induced cognitive dysfunction was significantly decreased by administration of the extract solution, in the passive avoidance task and inhibited brain acetylcholinesterase activity. These results suggest that bark extract of Pterocarpus santalinus can be better option for further studies on AD via their acetylcholinesterase inhibitory actions.

Keywords: Pterocarpus santalinus, cholinesterase inhibitor, passive avoidance, Alzheimer’s disease

Procedia PDF Downloads 231

Authors: Sarita Agarwal, Deepika Delsa Dean

Abstract:

Like the task of transferring photo images to artistic images, image-to-image translation aims to translate the data to the imitated data which belongs to the target domain. Neural Style Transfer and CycleGAN are two well-known deep learning architectures used for photo image-to-art image transfer. However, studies involving these two models concentrate on one-to-one domain translation, not one-to-multi domains translation. Our study tries to investigate deep learning architectures, which can be controlled to yield multiple artistic style translation only by adding a conditional vector. We have expanded CycleGAN and constructed Conditional CycleGAN for 5 kinds of categories translation. Our study found that the architecture inserting conditional vector into the middle layer of the Generator could output multiple artistic images.

Keywords: genetic counseling, FMR1 gene, fragile x-associated primary ovarian insufficiency, premutation

Procedia PDF Downloads 114

Authors: Danute Karcauskiene, Dalia Ambrazaitiene, Regina Skuodiene, Monika Vilkiene, Regina Repsiene, Ieva Jokubauskaite

Abstract:

The main ambition for nowadays agriculture is to get the economically effective yield and to secure the soil ecological sustainability. According to the effect on the main soil quality indexes, tillage systems may be separated into two types, conventional and conservation tillage. The goal of this study was to determine the impact of conservation and conventional primary soil tillage methods and soil fertility improvement measures on soil properties and crop productivity. Methods: The soil of the experimental site is Dystric Glossic Retisol (WRB 2014) with texture of sandy loam. The trial was established in 2003 in the experimental field of crop rotation of Vėžaičiai Branch of Lithuanian Research Centre for Agriculture and Forestry. Trial factors and treatments: factor A- primary soil tillage in (autumn): deep ploughing (20-25cm), shallow ploughing (10-12cm), shallow ploughless tillage (8-10cm); factor B – soil fertility improvement measures: plant residues, plant residues + straw, green manure 1st cut + straw, farmyard manure 40tha-1 + straw. The four - course crop rotation consisted of red clover, winter wheat, spring rape and spring barley with undersown. Results: The tillage had no statistically significant effect on topsoil (0-10 cm) pHKCl level, it was 5.5 - 5.7. During all experiment period, the highest soil pHKCl level (5.65) was in the shallow ploughless tillage. The organic fertilizers particularly the biomass of grass and farmyard manure had tendency to increase the soil pHKCl. The content of plant - available phosphorus and potassium significantly increase in the shallow ploughing compared with others tillage systems. The farmyard manure increases those elements in whole arable layer. The dissolved organic carbon concentration was significantly higher in the 0 - 10 cm soil layer in the shallow ploughless tillage compared with deep ploughing. After the incorporation of clover biomass and farmyard manure the concentration of dissolved organic carbon increased in the top soil layer. During all experiment period the largest amount of water stable aggregates was determined in the soil where the shallow ploughless tillage was applied. It was by 12% higher compared with deep ploughing. During all experiment time, the soil moisture was higher in the shallow ploughing and shallow ploughless tillage (9-27%) compared to deep ploughing. The lowest emission of CO2 was determined in the deep ploughing soil. The highest rate of CO2 emission was in shallow ploughless tillage. The addition of organic fertilisers had a tendency to increase the CO2 emission, but there was no statistically significant effect between the different types of organic fertilisers. The crop yield was larger in the deep ploughing soil compared to the shallow and shallow ploughless tillage.

Keywords: reduced tillage, soil structure, soil pH, biological activity, crop productivity

Procedia PDF Downloads 254

Authors: Pulari C. Milu Maria Anto

Abstract:

Background: Cognitive Rehabilitation/Retraining has been variously used in the research literature to represent non-pharmacological interventions that target the cognitive impairments with the goal of ameliorating cognitive function and functional behaviors to optimize the quality of life. Along with adult’s cognitive impairments, the need to address acquired cognitive impairments (due to any chronic illnesses like CHD - congenital heart diseases or ALL - Acute Lymphoblastic Leukemia) among child populations is inevitable. Also, it has to be emphasized as same we consider the cognitive impairments seen in the children having neurodevelopmental disorders. Methods: All published brain image studies (Hermann, B. et al,2002, Khalil, A. et al., 2004, Follin, C. et al, 2016, etc.) and studies emphasizing cognitive impairments in attention, memory, and/or executive function and behavioral aspects (Henkin, Y. et al,2007, Bellinger, D. C., & Newburger, J. W. (2010), Cheung, Y. T., et al,2016, that could be identified were reviewed. Based on a systematic review of the literature from (2000 -2021) different brain imaging studies, increased risk of neuropsychological and psychosocial impairments are briefly described. Clinical and research gap in the area is discussed. Results:30 papers, both Indian studies and foreign publications (Sage journals, Delhi psychiatry journal, Wiley Online Library, APA PsyNet, Springer, Elsevier, Developmental medicine, and child neurology), were identified. Conclusions: In India, a very limited number of brain imaging studies and neuropsychological studies have done by indicating the cognitive deficits of a child having or undergone chronic illness. None of the studies have emphasized the relevance nor the need of implementingCR among such children, even though its high time to address but still not established yet. The review of the current evidence is to bring out an insight among rehabilitation professionals in establishing a child specific CR and to publish new findings regarding the implementation of CR among such children. Also, this study will be an awareness on considering cognitive aspects of a child having acquired cognitive deficit (due to chronic illness), especially during their critical developmental period.

Keywords: cognitive rehabilitation, neuropsychological impairments, congenital heart diseases, acute lymphoblastic leukemia, epilepsy, and neuroplasticity

Procedia PDF Downloads 165

Authors: Chamika Chiran Perera, Dananjaya Perera, Chirath Dasanayake, Banuka Athuraliya

Abstract:

Male infertility is a major problem in the world, and it is a neglected and sensitive health issue in Sri Lanka. It can be determined by analyzing human semen samples. Sperm motility is one of many factors that can evaluate male’s fertility potential. In Sri Lanka, this analysis is performed manually. Manual methods are time consuming and depend on the person, but they are reliable and it can depend on the expert. Machine learning and deep learning technologies are currently being investigated to automate the spermatozoa motility analysis, and these methods are unreliable. These automatic methods tend to produce false positive results and false detection. Current automatic methods support different techniques, and some of them are very expensive. Due to the geographical variance in spermatozoa characteristics, current automatic methods are not reliable for motility analysis in Sri Lanka. The suggested system, DeepMotile, is to explore a method to analyze motility of human spermatozoa automatically and present it to the andrology laboratories to overcome current issues. DeepMotile is a novel deep learning method for analyzing spermatozoa motility parameters in the Sri Lankan population. To implement the current approach, Sri Lanka patient data were collected anonymously as a dataset, and glass slides were used as a low-cost technique to analyze semen samples. Current problem was identified as microscopic object detection and tackling the problem. YOLOv5 was customized and used as the object detector, and it achieved 94 % mAP (mean average precision), 86% Precision, and 90% Recall with the gathered dataset. StrongSORT was used as the object tracker, and it was validated with andrology experts due to the unavailability of annotated ground truth data. Furthermore, this research has identified many potential ways for further investigation, and andrology experts can use this system to analyze motility parameters with realistic accuracy.

Keywords: computer vision, deep learning, convolutional neural networks, multi-target tracking, microscopic object detection and tracking, male infertility detection, motility analysis of human spermatozoa

Procedia PDF Downloads 92

Authors: Hee Soo Kim, Ha Young Kyung

Abstract:

Happiness and pleasure are a result of dopamine, oxytocin, serotonin, and endorphin levels in the body. In order to increase the four neurochemical levels, it is important to associate daily activities with its corresponding neurochemical releases. This includes setting goals, maintaining social relationships, laughing frequently, and exercising regularly. The likelihood of experiencing happiness increases when all four neurochemicals are released at the optimal level. The achievement of happiness is important because it increases healthiness, productivity, and the ability to overcome adversity. To process emotions, electrical brain waves, brain structure, and neurochemicals must be analyzed. This research uses Chemcraft and Avogadro to determine the theoretical and chemical properties of the four neurochemical molecules. Each neurochemical molecule’s thermodynamic stability is calculated to observe the efficiency of the molecules. The study found that among dopamine, oxytocin, serotonin, alpha-, beta-, and gamma-endorphin, beta-endorphin has the lowest optimized energy of 388.510 kJ/mol. Beta-endorphin, a neurotransmitter involved in mitigating pain and stress, is the most thermodynamically stable and efficient molecule that is involved in the process of happiness. Through examining such properties of happiness neurotransmitters, the science of happiness is better understood.

Keywords: happiness, neurotransmitters, positive psychology, dopamine, oxytocin, serotonin, endorphins

Procedia PDF Downloads 145

Authors: Samiah Alammari, Nassim Ammour

Abstract:

When providing a massive number of tasks successively to a deep learning process, a good performance of the model requires preserving the previous tasks data to retrain the model for each upcoming classification. Otherwise, the model performs poorly due to the catastrophic forgetting phenomenon. To overcome this shortcoming, we developed a successful continual learning deep model for remote sensing hyperspectral image regions classification. The proposed neural network architecture encapsulates two trainable subnetworks. The first module adapts its weights by minimizing the discrimination error between the land-cover classes during the new task learning, and the second module tries to learn how to replicate the data of the previous tasks by discovering the latent data structure of the new task dataset. We conduct experiments on HSI dataset Indian Pines. The results confirm the capability of the proposed method.

Keywords: continual learning, data reconstruction, remote sensing, hyperspectral image segmentation

Procedia PDF Downloads 238

Authors: Salma A. El-Marasy, Rehab F. Abdel-Rahman, Reham M. Abd-Elsalam

Abstract:

The burden of stroke is intensely increasing worldwide. Brain injury following transient or permanent focal cerebral ischemia develops ischemic stroke as a consequence of a complex series of pathophysiological events. The aim of this study is to evaluate the possible neuroprotective effect of a dipeptidyl peptidase-4 inhibitor, vildagliptin, independent on its insulinotropic properties in non-diabetic rats subjected to cerebral ischemia. Anaesthetized Wistar rats were subjected to either left middle cerebral artery occlusion (MCAO) or sham operation followed by reperfusion after 30 min of MCAO. The other three groups were orally administered vildagliptin at 3 dose levels (2.5, 5, 10 mg/kg) for 3 successive weeks before subjected to left focal cerebral ischemia/reperfusion and till the end of the study. Neurological deficit scores and motor activity were assessed 24h following reperfusion. 48h following reperfusion, rats were euthanized and their left brain hemispheres were harvested and used in the biochemical, histopathological, and immunohistochemical investigations. Vildagliptin pretreatment improved neurological score deficit, locomotor activity and motor coordination in MCAO rats. Moreover, vildagliptin reduced malondialdehyde (MDA), elevated reduced glutathione (GSH), phosphotylinosital 3 kinase (PI3K), phosphorylated of protein kinase B (p-AKT), and mechanistic target of rapamycin (mTOR) brain contents in addition to reducing protein expression of caspase-3. Also, vildagliptin showed a dose-dependent attenuation in neuronal cell loss and histopathological alterations in MCAO rats. This study proves that vildagliptin exerted the neuroprotective effect in a dose-dependent manner as shown in amelioration of neuronal cell loss and histopathological damage in MCAO rats, which may be mediated by attenuating neuronal and motor deficits, it’s anti-oxidant property, activation of PI3K/AKT/mTOR pathway and its anti-apoptotic effect.

Keywords: caspase-3, cerebral ischemia, dipeptidyl peptidase-4 inhibitor, oxidative stress, PI3K/AKT/mTOR pathway, rats, vildagliptin

Procedia PDF Downloads 143

Authors: Beth Taylor, Kojima Mituaki, Atsushi Senda, Koji Morishita, Yasuhiro Otomo

Abstract:

Intestinal ischaemia-reperfusion injury drives systemic inflammation and organ failure following trauma/haemorrhagic shock (T/HS), through the release of pro-inflammatory mediators into the mesenteric lymph (ML). However, changes in the biological function of ML are not fully understood, and therefore, a specific model of intestinal ischaemia-reperfusion injury is required to obtain ML for the study of its biological function upon inflammatory cells. ML obtained from a model of intestinal ischaemia-reperfusion injury was used to assess biological function upon inflammatory cells and investigate changes in the biological function of individual ML components. An additional model was used to determine the effect of vagal nerve stimulation (VNS) upon biological function. Rat ML was obtained by mesenteric lymphatic duct cannulation before and after occlusion of the superior mesenteric artery (SMAO). ML was incubated with human polymorphonuclear neutrophils (PMNs), monocytes and lymphocytes, and the biological function of these cells was assessed. ML was then separated into supernatant, exosome and micro-vesicle components, and biological activity was compared in monocytes. A model with an additional VNS phase was developed, in which the right cervical vagal nerve was exposed and stimulated, and ML collected for comparison of biological function with the conventional model. The biological function of ML was altered by intestinal ischaemia-reperfusion injury, increasing PMN activation, monocyte activation, and lymphocyte apoptosis. Increased monocyte activation was only induced by the exosome component of ML, with no significant changes induced by the supernatant or micro-vesicle components. VNS partially attenuated monocyte activation, but no attenuation of PMN activation was observed. Intestinal ischaemia-reperfusion injury induces changes in the biological function of ML upon both innate and adaptive inflammatory cells, supporting the role of intestinal ischaemia-reperfusion injury in driving systemic inflammation following T/HS. The exosome component of ML appears to be critical to the transport of pro-inflammatory mediators in ML. VNS partially attenuates changes in innate inflammatory cell biological activity observed, presenting possibilities for future novel treatment development in multiple organ failure patients.

Keywords: exosomes, inflammation, intestinal ischaemia, mesenteric lymph, vagal stimulation

Procedia PDF Downloads 119

Authors: Mustafa M. Donma, Orkide Donma

Abstract:

The impact of metabolic syndrome (MetS) on cognition and functions of the brain is being investigated. Iron deficiency and deficiencies of B9 (folate) as well as B12 (cobalamin) vitamins are best-known nutritional anemias. They are associated with cognitive disorders and learning difficulties. The antidepressant effects of vitamin D are known and the deficiency state affects mental functions negatively. The aim of this study is to investigate possible correlations of MetS with serum brain-derived neurotrophic factor (BDNF), iron, folate, cobalamin and vitamin D in pediatric patients. 30 children, whose age- and sex-dependent body mass index (BMI) percentiles vary between 85 and 15, 60 morbid obese children with above 99^th percentiles constituted the study population. Anthropometric measurements were taken. BMI values were calculated. Age- and sex-dependent BMI percentile values were obtained using the appropriate tables prepared by the World Health Organization (WHO). Obesity classification was performed according to WHO criteria. Those with MetS were evaluated according to MetS criteria. Serum BDNF was determined by enzyme-linked immunosorbent assay. Serum folate was analyzed by an immunoassay analyzer. Serum cobalamin concentrations were measured using electrochemiluminescence immunoassay. Vitamin D status was determined by the measurement of 25-hydroxycholecalciferol [25-hydroxy vitamin D3, 25(OH)D] using high performance liquid chromatography. Statistical evaluations were performed using SPSS for Windows, version 16. The p values less than 0.05 were accepted as statistically significant. Although statistically insignificant, lower folate and cobalamin values were found in MO children compared to those observed for children with normal BMI. For iron and BDNF values, no alterations were detected among the groups. Significantly decreased vitamin D concentrations were noted in MO children with MetS in comparison with those in children with normal BMI (p ≤ 0.05). The positive correlation observed between iron and BDNF in normal-BMI group was not found in two MO groups. In THE MetS group, the partial correlation among iron, BDNF, folate, cobalamin, vitamin D controlling for waist circumference and BMI was r = -0.501; p ≤ 0.05. None was calculated in MO and normal BMI groups. In conclusion, vitamin D should also be considered during the assessment of pediatric MetS. Waist circumference and BMI should collectively be evaluated during the evaluation of MetS in children. Within this context, BDNF appears to be a key biochemical parameter during the examination of obesity degree in terms of mental functions, cognition and learning capacity. The association observed between iron and BDNF in children with normal BMI was not detected in MO groups possibly due to development of inflammation and other obesity-related pathologies. It was suggested that this finding may contribute to mental function impairments commonly observed among obese children.

Keywords: brain-derived neurotrophic factor, iron, vitamin B9, vitamin B12, vitamin D

Procedia PDF Downloads 102

Authors: Prakash Kumar Nayak, Avinash Kumar, Uma Dash, Kalpana Rayaguru

Abstract:

Soybean oil is one of the most widely consumed cooking oils, worldwide. Deep-fat frying of foods at higher temperatures adds unique flavour, golden brown colour and crispy texture to foods. But it brings in various changes like hydrolysis, oxidation, hydrogenation and thermal alteration to oil. The presence of Peroxide value (PV) is one of the most important factors affecting the quality of the deep-fat fried oil. Using bentonite as an adsorbent, the PV can be reduced, thereby improving the quality of the soybean oil. In this study, operating parameters like heating time of oil (10, 15, 20, 25 & 30 h), contact time ( 5, 10, 15, 20, 25 h) and concentration of adsorbent (0.25, 0.5, 0.75, 1.0 and 1.25 g/ 100 ml of oil) have been optimized by response surface methodology (RSM) considering percentage reduction of PV as a response. Adsorption data were analysed by fitting with Langmuir and Freundlich isotherm model. The results show that the Langmuir model shows the best fit compared to the Freundlich model. The adsorption process was also found to follow a pseudo-second-order kinetic model.

Keywords: bentonite, Langmuir isotherm, peroxide value, RSM, soybean oil

Procedia PDF Downloads 359

Authors: Yi-Feng Kao, Huey-Jine Chai, Chin-I Chang, Yi-Chen Chen, June-Ru Chen

Abstract:

Microglia is a macrophage that resides in brain, and overactive microglia may result in brain neuron damage or inflammation. In this study, the phospholipids was extracted from squid skin and manufactured into a liposome (SQ liposome) to mimic apoptotic body. We then evaluated anti-inflammatory effects of SQ liposome on mouse microglial cell line (BV-2) by lipopolysaccharide (LPS) induction. First, the major phospholipid constituents in the squid skin extract were including 46.2% of phosphatidylcholine, 18.4% of phosphatidylethanolamine, 7.7% of phosphatidylserine, 3.5% of phosphatidylinositol, 4.9% of Lysophosphatidylcholine and 19.3% of other phospholipids by HPLC-UV analysis. The contents of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) in the squid skin extract were 11.8 and 28.7%, respectively. The microscopic images showed that microglia cells can engulf apoptotic cells or SQ-liposome. In cell based studies, there was no cytotoxicity to BV-2 as the concentration of SQ-liposome was less than 2.5 mg/mL. The LPS induced pro-inflammatory cytokines, including tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6), were significant suppressed (P < 0.05) by pretreated 0.03~2.5mg/ml SQ liposome. Oppositely, the anti-inflammatory cytokines transforming growth factor-beta (TGF-β) and interleukin-10 (IL-10) secretion were enhanced (P < 0.05). The results suggested that SQ-liposome possess anti-inflammatory properties on BV-2 and may be a good strategy for against neuro-inflammatory disease.

Keywords: apoptotic mimicry, neuroinflammation, microglia, squid processing by-products

Procedia PDF Downloads 468

Authors: Chuluundorj Begz

Abstract:

This paper presents the dynamic, psycho-cognitive approach to study of human verbal thinking on the basis of typologically different languages /as a Mongolian, English and Russian/. Topological equivalence in verbal communication serves as a basis of Universality of mental structures and therefore deep structures. Mechanism of verbal thinking consisted at the deep level of basic concepts, rules for integration and classification, neural networks of vocabulary. In neuro cognitive study of language, neural architecture and neuro psychological mechanism of verbal cognition are basis of a vector-based modeling. Verbal perception and interpretation of the infinite set of meanings and propositions in mental continuum can be modeled by applying tensor methods. Euclidean and non-Euclidean spaces are applied for a description of human semantic vocabulary and high order structures.

Keywords: Euclidean spaces, isomorphism and homomorphism, mental lexicon, mental mapping, semantic memory, verbal cognition, vector space

Procedia PDF Downloads 507

Authors: Jacob Green, Cecilia Cabrera, Maximilian Jakobs, Andrea Dimitracopoulos, Mark van der Wilk, Ryan Greenhalgh

Abstract:

Predicting drug properties is key in drug discovery to enable de-risking of assets before expensive clinical trials and to find highly active compounds faster. Interest from the machine learning community has led to the release of a variety of benchmark datasets and proposed methods. However, it remains unclear for practitioners which method or approach is most suitable, as different papers benchmark on different datasets and methods, leading to varying conclusions that are not easily compared. Our large-scale empirical study links together numerous earlier works on different datasets and methods, thus offering a comprehensive overview of the existing property classes, datasets, and their interactions with different methods. We emphasise the importance of uncertainty quantification and the time and, therefore, cost of applying these methods in the drug development decision-making cycle. To the best of the author's knowledge, it has been observed that the optimal approach varies depending on the dataset and that engineered features with classical machine learning methods often outperform deep learning. Specifically, QSAR datasets are typically best analysed with classical methods such as Gaussian Processes, while ADMET datasets are sometimes better described by Trees or deep learning methods such as Graph Neural Networks or language models. Our work highlights that practitioners do not yet have a straightforward, black-box procedure to rely on and sets a precedent for creating practitioner-relevant benchmarks. Deep learning approaches must be proven on these benchmarks to become the practical method of choice in drug property prediction.

Keywords: activity (QSAR), ADMET, classical methods, drug property prediction, empirical study, machine learning

Procedia PDF Downloads 59

Authors: Sherri Franklin-Guy

Abstract:

The emergence of language in young children has been well-documented and play-based activities that support its continued development have been utilized in the clinic-based setting. Speech-language pathologists have long used such activities to stimulate the production of language in children with speech and language disorders via modeling and elicitation tasks. This presentation will examine the importance of play in the development of language in young children, including social and pragmatic communication. Implications for clinicians and educators will be discussed.

Keywords: language development, language stimulation, play-based activities, symbolic play

Procedia PDF Downloads 226

Authors: Alexandra K. Diem, Giles Richardson, Roxana O. Carare, Neil W. Bressloff

Abstract:

Alzheimer's disease (AD) is the most common form of dementia and although it has been researched for over 100 years, there is still no cure or preventive medication. Its onset and progression is closely related to the accumulation of the neuronal metabolite Aβ. This raises the question of how metabolites and waste products are eliminated from the brain as the brain does not have a traditional lymphatic system. In recent years the rapid uptake of Aβ into cerebral artery walls and its clearance along those arteries towards the lymph nodes in the neck has been suggested and confirmed in mice studies, which has led to the hypothesis that interstitial fluid (ISF), in the basement membranes in the walls of cerebral arteries, provides the pathways for the lymphatic drainage of Aβ. This mechanism, however, requires a net reverse flow of ISF inside the blood vessel wall compared to the blood flow and the driving forces for such a mechanism remain unknown. While possible driving mechanisms have been studied using mathematical models in the past, a mechanism for net reverse flow has not been discovered yet. Here, we aim to address the question of the driving force of this reverse lymphatic drainage of Aβ (also called perivascular drainage) by using multi-scale numerical and analytical modelling. The numerical simulation software COMSOL Multiphysics 4.4 is used to develop a fluid-structure interaction model of a cerebral artery, which models blood flow and displacements in the artery wall due to blood pressure changes. An analytical model of a layer of basement membrane inside the wall governs the flow of ISF and, therefore, solute drainage based on the pressure changes and wall displacements obtained from the cerebral artery model. The findings suggest that an active role in facilitating a reverse flow is played by the components of the basement membrane and that stiffening of the artery wall during age is a major risk factor for the impairment of brain lymphatics. Additionally, our model supports the hypothesis of a close association between cerebrovascular diseases and the failure of perivascular drainage.

Keywords: Alzheimer's disease, artery wall mechanics, cerebral blood flow, cerebral lymphatics

Procedia PDF Downloads 514

Authors: Naim Izet Kajtazi

Abstract:

Context: The third ventricle colloid cyst (CC) is a benign growth usually located in the third ventricle and can cause various neurological symptoms, including sudden death. Modern surgical interventions may still result in a wide range of complications and cerebral venous thrombosis (CVT) is among them. Process: A 38-year-old female with an existing diagnosis of diabetes mellitus (DM) and hypothyroidism and a six-month history of headaches, blurred vision, and vomiting presented to our clinic three days after the headaches became excessively severe. Neurological examination on admission revealed bilateral papilledema without any associated focal neurological deficits. Brain computed tomography (CT) and magnetic resonance imaging (MRI) confirmed the presence of a third ventricle colloid cyst and associated non-communicating hydrocephalus involving the lateral ventricles. As a result, the patient underwent emergency bilateral external ventricular drainage (EVD) insertion followed by a third ventricular CC excision under neuronavigation through a right frontal craniotomy. Twelve days post-operatively, the patient developed further headaches, followed by a generalized tonic-clonic seizure that led to no postictal neurological deficits. Nonetheless, computed tomography venography of the brain revealed extensive thrombosis of the superior sagittal sinus, inferior sagittal sinus, right sigmoid sinus, and right internal jugular vein. A newly diagnosed CVT was treated with intravenous heparin. The patient was discharged with warfarin, which was discontinued after 12 months. Ten years after her illness, she remained stable and free from any neurological deficits but still suffered from mild chronic headaches. Outcome: Ten years after her illness, she remained stable and free from any neurological deficits but still suffered from mild chronic headaches. Relevance: A preoperative venous study should be performed in all cases to gain a better understanding of the venous anatomy. We advocate meticulous microsurgical techniques to protect the venous system surrounding the foramen of Monro and reduce the amount of retraction during surgery.

Keywords: CVT, seizures, third ventricle colloid cyst, MRI of brain

Procedia PDF Downloads 55

Authors: F. Polito, M. Cucinotta, A. Conti, C. Lo Giudice, C. Tomasello, F. Angileri, D. La Torre, M. Aguennouz

Abstract:

Glioblastoma multiforme (GBM) is the most aggressive form of brain tumors, with an extremely poor prognosis. Telomeres lenght is associated with tumor progression in several type of human cancers and telomere elongation is a common molecular feature of advanced malignancies. Among the telomeric shelterin proteins PTOP is required for telomeric protein complex assembly, telomerase recruitment and activity, and telomere length regulation through a PTOP-telomerase interaction. Previous studies suggest that PTOP upregulation is involved in radioresistance and telomere lengthening in colorectal cancer cells. Moreover, in human osteosarcoma cells PTOP deletion led to telomere shortening, increased apoptosis and radiation sensitivity enhancement. However, to date, little is known about the role of PTOP in progression of glioma cancers. In light of this background aim of the study is to investigate the expression of PTOP in different grades of human glioma and its correlation with the pathological grade of gliomas, grades of malignancy, proliferative activity and apoptosis. Fifteen Low Grade Astrocytomas (LGA), 18 Anaplastic Astrocytomas (AA) and 26 Glioblastoma Multiforme (GBM) samples were analyzed. Three samples of normal brain tissue (NBT) were used as controls. The expression levels of PTOP, h-TERT, BIRC1 and cyclin D1 were determined by real time PCR and/or western blot. Results obtained shows that PTOP expression in glioma tissues is tightly correlated with clinical grade ( p < 0.01 ). No correlation was found between PTOP expression and other clinicopathologic parameters. The expression of PTOP was positively correlated with the expression of hTERT and TERF1. Furthermore PTOP positively correlates with cyclin D1 and negatively correlates with the expression of BIRC1. Our findings indicate that PTOP might play key role in the progression of glioma regulating telomerase activity and likely through regulation of cell cycle and apoptosis. In conclusion results obtained prompted us to speculate that PTOP might represents a potential molecular bio marker and a therapeutic target for the treatment of glioblastoma tumors.

Keywords: glioblastoma, PTOP, telomere, brain tumors

Procedia PDF Downloads 328

Authors: Wann-Yun Shieh, Chin-Man Wang, Ya-Cheng Shieh

Abstract:

This study aims to develop an objective scoring system to evaluate hand-eye coordination and hand dexterity for Parkinson’s disease. This system contains three boards, and each of them is implemented with the sensors to sense a user’s finger operations. The operations include the peg test, the block test, and the blind block test. A user has to use the vision, hearing, and tactile abilities to finish these operations, and the board will record the results automatically. These results can help the physicians to evaluate a user’s reaction, coordination, dexterity function. The results will be collected to a cloud database for further analysis and statistics. A researcher can use this system to obtain systematic, graphic reports for an individual or a group of users. Particularly, a deep learning model is developed to learn the features of the data from different users. This model will help the physicians to assess the Parkinson’s disease symptoms by a more intellective algorithm.

Keywords: deep learning, hand-eye coordination, reaction, hand dexterity

Procedia PDF Downloads 51

Authors: Tomás Sobrino, Lara García-Varela, Marta Aramburu-Núñez, Mónica Castro, Noemí Gómez-Lado, Mariña Rodríguez-Arrizabalaga, Antía Custodia, Juan Manuel Pías-Peleteiro, José Manuel Aldrey, Daniel Romaus-Sanjurjo, Ángeles Almeida, Pablo Aguiar, Alberto Ouro

Abstract:

Introduction: Alzheimer’s disease (AD) and other tauopathies are primary causes of dementia, causing progressive cognitive deterioration that entails serious repercussions for the patients' performance of daily tasks. Currently, there is no effective approach for the early diagnosis and treatment of AD and tauopathies. This study suggests a theragnostic approach based on the importance of phosphorylated tau protein (p-Tau) in the early pathophysiological processes of AD. We have developed a novel theragnostic monoclonal antibody (mAb) to provide both diagnostic and therapeutic effects. Methods/Results: We have developed a p-Tau mAb, which was doped with deferoxamine for radiolabeling with Zirconium-89 (89Zr) for PET imaging, as well as fluorescence dies for immunofluorescence assays. The p-Tau mAb was evaluated in vitro for toxicity by MTT assay, LDH activity, propidium iodide/Annexin V assay, caspase-3, and mitochondrial membrane potential (MMP) assay in both mouse endothelial cell line (bEnd.3) and cortical primary neurons cell cultures. Importantly, non-toxic effects (up to concentrations of p-Tau mAb greater than 100 ug/mL) were detected. In vivo experiments in the tauopathy model mice (PS19) show that the 89Zr-pTau-mAb and 89Zr-Fragments-pTau-mAb are stable in circulation for up to 10 days without toxic effects. However, only less than 0.2% reached the brain, so further strategies have to be designed for crossing the Brain-Blood-Barrier (BBB). Moreover, an intraparenchymal treatment strategy was carried out. The PS19 mice were operated to implement osmotic pumps (Alzet 1004) at two different times, at 4 and 7 months, to stimulate the controlled release for one month each of the B6 antibody or the IgG1 control antibody. We demonstrated that B6-treated mice maintained their motor and memory abilities significantly compared with IgG1 treatment. In addition, we observed a significant reduction in p-Tau deposits in the brain. Conclusions /Discussion: A theragnostic pTau-mAb was developed. Moreover, we demonstrated that our p-Tau mAb recognizes very-early pathology forms of p-Tau by non-invasive techniques, such as PET. In addition, p-Tau mAb has non-toxic effects, both in vitro and in vivo. Although the p-Tau mAb is stable in circulation, only 0.2% achieve the brain. However, direct intraventricular treatment significantly reduces cognitive impairment in Alzheimer's animal models, as well as the accumulation of toxic p-Tau species.

Keywords: alzheimer's disease, theragnosis, tau, PET, immunotherapy, tauopathies

Procedia PDF Downloads 56

Authors: Airy Huang, Fuji Foo, Aries Prasetya Wibowo

Abstract:

In recent years, the focus of Natural Language Processing (NLP) development has been gradually shifting from the semantics-based approach to deep learning one, which performs faster with fewer resources. Although it performs well in many applications, the deep learning approach, due to the lack of semantics understanding, has difficulties in noticing and expressing a novel business case with a pre-defined scope. In order to meet the requirements of specific robotic services, deep learning approach is very labor-intensive and time consuming. It is very difficult to improve the capabilities of conversational AI in a short time, and it is even more difficult to self-learn from experiences to deliver the same service in a better way. In this paper, we present an adaptive conversational AI algorithm that combines both semantic knowledge and deep learning to address this issue by learning new business cases through conversations. After self-learning from experience, the robot adapts to the business cases originally out of scope. The idea is to build new or extended robotic services in a systematic and fast-training manner with self-configured programs and constructed dialog flows. For every cycle in which a chat bot (conversational AI) delivers a given set of business cases, it is trapped to self-measure its performance and rethink every unknown dialog flows to improve the service by retraining with those new business cases. If the training process reaches a bottleneck and incurs some difficulties, human personnel will be informed of further instructions. He or she may retrain the chat bot with newly configured programs, or new dialog flows for new services. One approach employs semantics analysis to learn the dialogues for new business cases and then establish the necessary ontology for the new service. With the newly learned programs, it completes the understanding of the reaction behavior and finally uses dialog flows to connect all the understanding results and programs, achieving the goal of self-learning process. We have developed a chat bot service mounted on a kiosk, with a camera for facial recognition and a directional microphone array for voice capture. The chat bot serves as a concierge with polite conversation for visitors. As a proof of concept. We have demonstrated to complete 90% of reception services with limited self-learning capability.

Keywords: conversational AI, chatbot, dialog management, semantic analysis

Procedia PDF Downloads 124

Authors: Thinh Ngo, Paul Rad, Brian Kelley

Abstract:

Noise estimation is essential in today wireless systems for power control, adaptive modulation, interference suppression and quality of service. Deep learning (DL) has already been applied in the physical layer for modulation and signal classifications. Unacceptably low accuracy of less than 50% is found to undermine traditional application of DL classification for SNR prediction. In this paper, we use divide-and-conquer algorithm and classifier fusion method to simplify SNR classification and therefore enhances DL learning and prediction. Specifically, multiple CNNs are used for classification rather than a single CNN. Each CNN performs a binary classification of a single SNR with two labels: less than, greater than or equal. Together, multiple CNNs are combined to effectively classify over a range of SNR values from −20 ≤ SNR ≤ 32 dB.We use pre-trained CNNs to predict SNR over a wide range of joint channel parameters including multiple Doppler shifts (0, 60, 120 Hz), power-delay profiles, and signal-modulation types (QPSK,16QAM,64-QAM). The approach achieves individual SNR prediction accuracy of 92%, composite accuracy of 70% and prediction convergence one order of magnitude faster than that of traditional estimation.

Keywords: classification, CNN, deep learning, prediction, SNR

Procedia PDF Downloads 124

Authors: Alexander Guzman Urbina, Atsushi Aoyama

Abstract:

The sustainability of traditional technologies employed in energy and chemical infrastructure brings a big challenge for our society. Making decisions related with safety of industrial infrastructure, the values of accidental risk are becoming relevant points for discussion. However, the challenge is the reliability of the models employed to get the risk data. Such models usually involve large number of variables and with large amounts of uncertainty. The most efficient techniques to overcome those problems are built using Artificial Intelligence (AI), and more specifically using hybrid systems such as Neuro-Fuzzy algorithms. Therefore, this paper aims to introduce a hybrid algorithm for risk assessment trained using near-miss accident data. As mentioned above the sustainability of traditional technologies related with energy and chemical infrastructure constitutes one of the major challenges that today’s societies and firms are facing. Besides that, the adaptation of those technologies to the effects of the climate change in sensible environments represents a critical concern for safety and risk management. Regarding this issue argue that social consequences of catastrophic risks are increasing rapidly, due mainly to the concentration of people and energy infrastructure in hazard-prone areas, aggravated by the lack of knowledge about the risks. Additional to the social consequences described above, and considering the industrial sector as critical infrastructure due to its large impact to the economy in case of a failure the relevance of industrial safety has become a critical issue for the current society. Then, regarding the safety concern, pipeline operators and regulators have been performing risk assessments in attempts to evaluate accurately probabilities of failure of the infrastructure, and consequences associated with those failures. However, estimating accidental risks in critical infrastructure involves a substantial effort and costs due to number of variables involved, complexity and lack of information. Therefore, this paper aims to introduce a well trained algorithm for risk assessment using deep learning, which could be capable to deal efficiently with the complexity and uncertainty. The advantage point of the deep learning using near-miss accidents data is that it could be employed in risk assessment as an efficient engineering tool to treat the uncertainty of the risk values in complex environments. The basic idea of using a Near-Miss Deep Learning Approach for Neuro-Fuzzy Risk Assessment in Pipelines is focused in the objective of improve the validity of the risk values learning from near-miss accidents and imitating the human expertise scoring risks and setting tolerance levels. In summary, the method of Deep Learning for Neuro-Fuzzy Risk Assessment involves a regression analysis called group method of data handling (GMDH), which consists in the determination of the optimal configuration of the risk assessment model and its parameters employing polynomial theory.

Keywords: deep learning, risk assessment, neuro fuzzy, pipelines

Procedia PDF Downloads 283

Authors: Aneta Daniel

Abstract:

The subject of the study is the exploration of the origins and dynamics of the development of language studies, which have been labelled as neurolinguistics. It is worth mentioning that the origins of neurolinguistics are to be found in the research conducted by German scientists before the Second World War in Breslau Universität (presently Wroclaw). The dominant figure in these studies was professor Carl Wernicke, whose students continued and creatively developed projects of their master within this area. Professor Carl Wernicke, a German physician, anatomist, psychiatrist, and neuropathologist, is primarily known for his influential research on aphasia. His research, as well as those conducted by professor Paul Broca, has led to breakthroughs in the location of brain functions, particularly speech. Years later the theses of the pioneers of cognitive neurology (Carl Wernicke and Paul Broca) were developed by other neurolinguists. The main objective of the investigation is the reconstruction of the group of scientists –the students of Carl Wernicke– who contributed to the development of neurolinguistics. The scholars were mainly neurologists and psychiatrists and dealt with the branch of science that had not been named neurolinguistics at that time. The profiles of the scholars will be analysed and presented as the members of the group of researchers who have contributed to the breakthroughs in psychology and neuroscience. The research material consists of archival records documenting the research of professor Carl Wernicke and the researchers from Breslau (presently Wroclaw) which is one of the fastest growing cities in Europe. In 1870, when Carl Wernicke became the medical doctor, Breslau was full of cultural events: festivals and circus shows were held in the city center. Today we can come back to these events due to 'Breslauer Zeitung (1870)', which precisely describes all the events that took place on particular days. It is worth noting that those were the beginnings of antisemitism in Breslau. Many theses and articles that have survived in the libraries in Wroclaw and all over the world contribute to the development of neuroscience. The history of research on the brain and speech analysis, including the history of psychology and neuroscience, areas from which neurolinguistics is derived, will be presented.

Keywords: Aphasia, brain injury, Carl Wernicke, language, neurolinguistics

Procedia PDF Downloads 372

Authors: Sonia Boscenco, Zihan Wang, Euclides José de Mendoça Filho, João Paulo Hoppe, Irina Pokhvisneva, Geoffrey B.C. Hall, Michael J. Meaney, Patricia Pelufo Silveira

Abstract:

Entropy can be described as a measure of the number of states of a system, and when used in the context of physiological time-based signals, it serves as a measure of complexity. In functional connectivity data, entropy can account for the moment-to-moment variability that is neglected in traditional functional magnetic resonance imaging (fMRI) analyses. While brain fMRI resting state entropy has been associated with some pathological conditions like schizophrenia, no investigations have explored the association between brain entropy measures and individual differences in child behavior in healthy children. We describe a novel exploratory approach to evaluate brain fMRI resting state data in two child cohorts, and MAVAN (N=54, 4.5 years, 48% males) and GUSTO (N = 206, 4.5 years, 48% males) and its associations to child behavior, that can be used in future research in the context of child exposures and long-term health. Following rs-fMRI data pre-processing and Shannon entropy calculation across 32 network regions of interest to acquire 496 unique functional connections, partial correlation coefficient analysis adjusted for sex was performed to identify associations between entropy data and Strengths and Difficulties questionnaire in MAVAN and Child Behavior Checklist domains in GUSTO. Significance was set at p < 0.01, and we found eight significant associations in GUSTO. Negative associations were found between two frontoparietal regions and cerebellar posterior and oppositional defiant problems, (r = -0.212, p = 0.006) and (r = -0.200, p = 0.009). Positive associations were identified between somatic complaints and four default mode connections: salience insula (r = 0.202, p < 0.01), dorsal attention intraparietal sulcus (r = 0.231, p = 0.003), language inferior frontal gyrus (r = 0.207, p = 0.008) and language posterior superior temporal gyrus (r = 0.210, p = 0.008). Positive associations were also found between insula and frontoparietal connection and attention deficit / hyperactivity problems (r = 0.200, p < 0.01), and insula – default mode connection and pervasive developmental problems (r = 0.210, p = 0.007). In MAVAN, ten significant associations were identified. Two positive associations were found = with prosocial scores: the salience prefrontal cortex and dorsal attention connection (r = 0.474, p = 0.005) and the salience supramarginal gyrus and dorsal attention intraparietal sulcus (r = 0.447, p = 0.008). The insula and prefrontal connection were negatively associated with peer problems (r = -0.437, p < 0.01). Conduct problems were negatively associated with six separate connections, the left salience insula and right salience insula (r = -0.449, p = 0.008), left salience insula and right salience supramarginal gyrus (r = -0.512, p = 0.002), the default mode and visual network (r = -0.444, p = 0.009), dorsal attention and language network (r = -0.490, p = 0.003), and default mode and posterior parietal cortex (r = -0.546, p = 0.001). Entropy measures of resting state functional connectivity can be used to identify individual differences in brain function that are correlated with variation in behavioral problems in healthy children. Further studies applying this marker into the context of environmental exposures are warranted.

Keywords: child behaviour, functional connectivity, imaging, Shannon entropy

Procedia PDF Downloads 189

Authors: Gustavo Axel Elizalde-Velázquez

Abstract:

Metformin is the first line of oral therapy to treat type II diabetes and is also employed as a treatment for other indications, such as polycystic ovary syndrome, cancer, and COVID-19. Recent data suggest it is the aspirin of the 21st century due to its antioxidant and anti-aging effects. However, increasingly current articles indicate its long-term consumption generates mitochondrial impairment. Up to date, it is known metformin increases the biogenesis of Alzheimer's amyloid peptides via up-regulating BACE1 transcription, but further information related to brain damage after its consumption is missing. Bearing in mind the above, this work aimed to establish whether or not chronic exposure to metformin may alter swimming behavior and induce neurotoxicity in Danio rerio adults. For this purpose, 250 Danio rerio grown-ups were assigned to six tanks of 50 L of capacity. Four of the six systems contained 50 fish, while the remaining two had 25 fish (≈1 male:1 female ratio). Every system with 50 fish was allocated one of the three metformin treatment concentrations (1, 20, and 40 μg/L), with one system as the control treatment. Systems with 25 fish, on the other hand, were used as positive controls for acetylcholinesterase (10 μg/L of Atrazine) and oxidative stress (3 μg/L of Atrazine). After four months of exposure, a mean of 32 fish (S.D. ± 2) per group of MET treatment survived, which were used for the evaluation of behavior with the Novel Tank test. Moreover, after the behavioral assessment, we aimed to collect the blood and brains of all fish from all treatment groups. For blood collection, fish were anesthetized with an MS-222 solution (150 mg/L), while for brain gathering, fish were euthanized using the hypothermic shock method (2–4 °C). Blood was employed to determine CASP3 activity and the percentage of apoptotic cells with the TUNEL assay, and brains were used to evaluate acetylcholinesterase activity, oxidative damage, and gene expression. After chronic exposure, MET-exposed fish exhibited less swimming activity when compared to control fish. Moreover, compared with the control group, MET significantly inhibited the activity of AChE and induced oxidative damage in the brain of fish. Concerning gene expression, MET significantly upregulated the expression of Nrf1, Nrf2, BAX, p53, BACE1, APP, PSEN1, and downregulated CASP3 and CASP9. Although MET did not overexpress the CASP3 gene, we saw a meaningful rise in the activity of this enzyme in the blood of fish exposed to MET compared to the control group, which we then confirmed by a high number of apoptotic cells in the TUNEL assay. To the best of our understanding, this is the first study that delivers evidence of oxidative impairment, apoptosis, AChE alteration, and overexpression of B- amyloid-related genes in the brain of fish exposed to metformin.

Keywords: AChE inhibition, CASP3 activity, NovelTank test, oxidative damage, TUNEL assay

Procedia PDF Downloads 74

Authors: Habib Alah Dadgar, Nasim Norouzbeigi

Abstract:

The precise definition margin of high and low-grade gliomas is crucial for treatment. We aimed to assess the feasibility of assessment of the resection legions with post-operative positron emission tomography (PET) using [18F]O-(2-[18F]-fluoroethyl)-L-tyrosine ([18F]FET). Four patients with the suspicion of high and low-grade were enrolled. Patients underwent post-operative [18F]FET-PET, pre-operative magnetic resonance imaging (MRI) and CT for clinical evaluations. In our study, three patients had negative response to recurrence and progression and one patient indicated positive response after surgery. [18F]FET-PET revealed a legion of increased radiotracer uptake in the dura in the craniotomy site for patient 1. Corresponding to the patient history, the study was negative for recurrence of brain tumor. For patient 2, there was a lesion in the right parieto-temporal with slightly increased uptake in its posterior part with SUVmax = 3.79, so the study was negative for recurrence evaluation. In patient 3 there was no abnormal uptake with negative result for recurrence of brain tumor. Intense radiotracer uptake in the left parietal lobe where in the MRI there was a lesion with no change in enhancement in the post-contrast image is indicated in patient 4. Assessment of the resection legions in high and low-grade gliomas with [18F]FET-PET seems to be useful.

Keywords: FET-PET, CT, glioma, MRI

Procedia PDF Downloads 187

Authors: Khalida Douibi, Rodrigo Balp, Solène Le Bars

Abstract:

In the medical field, applications related to human experiments are frequently linked to reduced samples size, which makes the training of machine learning models quite sensitive and therefore not very robust nor generalizable. This is notably the case in Brain-Computer Interface (BCI) studies, where the sample size rarely exceeds 20 subjects or a few number of trials. To address this problem, several resampling approaches are often used during the data preparation phase, which is an overly critical step in a data science analysis process. One of the naive approaches that is usually applied by data scientists consists in the transformation of the entire database before the resampling phase. However, this can cause model’ s performance to be incorrectly estimated when making predictions on unseen data. In this paper, we explored the effect of data leakage observed during our BCI experiments for device control through the real-time classification of SSVEPs (Steady State Visually Evoked Potentials). We also studied potential ways to ensure optimal validation of the classifiers during the calibration phase to avoid overfitting. The results show that the scaling step is crucial for some algorithms, and it should be applied after the resampling phase to avoid data leackage and improve results.

Keywords: data leackage, data science, machine learning, SSVEP, BCI, overfitting

Procedia PDF Downloads 140