NUST Institutions Library Catalogue Search for 'an:119509'

Early Detection & Stage Classification of Parkinson’s Disease using Deep Learning /

By Zeeshan, Muhammad Muzzamil . . 38p. ; 30cm..

Place Hold on Early Detection & Stage Classification of Parkinson’s Disease using Deep Learning /

Ciprofloxacin-Gold Nanoparticles Loaded Polymeric Membrane as a Novel Antibacterial Wound Dressing against Diabetic Foot Ulcer Infection /

By Malik, Hadiya . . 77p. 30cm.

Place Hold on Ciprofloxacin-Gold Nanoparticles Loaded Polymeric Membrane as a Novel Antibacterial Wound Dressing against Diabetic Foot Ulcer Infection /

Design and Development of a 3D Bio-Printer /

By Hassan, Muhammad. . 57p. ; 30cm..

Place Hold on Design and Development of a 3D Bio-Printer /

White Matter Multiple Sclerosis Lesion Segmentation Under Distributional Shifts /

By Haider, Ali . . 37p. ; 30cm..

Place Hold on White Matter Multiple Sclerosis Lesion Segmentation Under Distributional Shifts /

An Augmentative Phonation and Articulation System using Advanced Signal Processing Techniques as an Alternative Communication Device /

By Shafiq, Uzma . . 50p. ; 30cm..

Place Hold on An Augmentative Phonation and Articulation System using Advanced Signal Processing Techniques as an Alternative Communication Device /

Comparison And Correlation of Hamstring Reinforcing Exercise With Sprint In terms of Muscles Activity And Force Production /

By Habib, Adnan . . 44p. 30cm.

Place Hold on Comparison And Correlation of Hamstring Reinforcing Exercise With Sprint In terms of Muscles Activity And Force Production /

Optimizing Feature Reduction and Selection Techniques for Surface Electromyography /

By Khan, Affaf . . 65p. 30cm.

Place Hold on Optimizing Feature Reduction and Selection Techniques for Surface Electromyography /

EMG Feature Reduction Technique For Optimal Accuracies /

By Abbas, Usman . . 94p. 30cm.

Place Hold on EMG Feature Reduction Technique For Optimal Accuracies /

Enhancing Gait Symmetry: The Impact of Prolonged Sitting and the Efficacy of Flexibility Exercises /

By Khan, Habib Ullah. . 62p. , Due to advancement in technology, we have seen a shift in the modern work environment. A notable surge has been seen in the percentage of people with jobs requiring prolonged sitting. There is a strong link between sedentary behavior and health risks down the road. Gait of a person is also linked to sedentary behavior which is believed to be significantly affected by prolonged inactivity. Gait asymmetry can be a very good indicator of overall physical health and that is why it used extensively to assess the physical health of subjects. Due to prolonged sitting, tightening of certain muscles, reduction in core strength and reduction in joint flexibility can lead to various Gait abnormalities. Our study is designed to compare the Gait asymmetry of physically active individuals with those of inactive individuals. We took into account various gait related features to find out the gait asymmetry between the right and left limb. We then also assessed the impact of exercise on gait symmetry and introduce flexibility training to the Sedentary group. Sedentary group performed Flexibility Training for 4 weeks every day. The exercises were performed thrice every day for 30 minutes. Results were substantial in supporting the link between sedentary behavior and increased gait asymmetry. Results also showed improvements in gait symmetry using flexibility training exercises. 30cm.

Place Hold on Enhancing Gait Symmetry: The Impact of Prolonged Sitting and the Efficacy of Flexibility Exercises /

Design and Development of Visual Inspection System of the stent using Image Processing Technique /

By Ahmed, Umair . . 80p. 30cm.

Place Hold on Design and Development of Visual Inspection System of the stent using Image Processing Technique /

Effect of load and size on wear depth per cycle in Total Knee Replacement (TKR) /

By Ahmed, Hajira. . 97p. , The sole objective of this research thesis is to evaluate the effect of size and load on the wear per cycle in prosthetic knee joints. Prosthetic knee joints are abundantly used in the medical field to cater the deteriorating natural knees and help maintain a better standard of living. The prosthetic knees come in seven universal sizes and each one acts differently under same load conditions. Initially, seven different models developed on SolidWorks constituting the femoral and disk part. The parts were mated so that the femoral component can slide on the disk. Then, the wear analysis was carried out using Abaqus. Firstly, materials were assigned to the two components i.e., Cobalt Chromium Molybdenum Alloy (CoCrMo) to the femoral component and Ultra High Molecular Weight Polyethylene Disk (UHMWPE) to the polyethylene disk. Meshing was done and boundary conditions were defined to execute the whole analysis. The use of this method resulted in the calculation of contact stresses on the area of contact between the two parts. The contact stresses were calculated at 700N, 950N and 1200N for each size. The contact stresses were maximum at the edges due to small area concentration and minimum at the center attributing to the large area. Moreover, the maximum contact stress was utilized to evaluate the wear depth by utilizing Archard’s wear law. The wear depth was calculated for one gait cycle. The wear depths (𝑊d) for size 1 against loads of 700 N, 950 N and 1200 N were 26.131 nm/cycle, 31.706 nm/cycle and 37.253 nm/cycle respectively. The 𝑊d for size 5 against loads of 700 N, 950 N and 1200 N were 140.27 nm/cycle, 156.14 nm/cycle and 171.45 nm/cycle respectively. The 𝑊d for size 6 against loads of 700 N, 950 N and 1200 N were 23.060 nm/cycle, 28.193 nm/cycle and 33.307 nm/cycle respectively. Similarly, the wear depth was calculated for the remaining sizes. The results indicate that the maximum wear occurs in size 5 and the minimum wear occurs in size 6. The incorporation of Cobalt Chromium Molybdenum Alloy (CoCrMo) and Ultra High Molecular Weight Polyethylene Disk (UHMWPE) has greatly reduced the wear debris per cycle and allowed the knee joint to work properly which are quite evident from the results. 30cm.

Place Hold on Effect of load and size on wear depth per cycle in Total Knee Replacement (TKR) /

Design of Passive Controlled Iot Based Smart Cpm for Lower Limb /

By Shahnaz, Iram . . 42p. , Muscle strength and range of motion in the knee joint are impacted following total knee replacement and anterior cruciate ligament surgery. As a result, the goal of this research is to design a rehabilitative device to improve knee joint muscle strength and range of motion. The global population of people with mobility disorders is growing, which increases the demand for rehabilitation and physiotherapy devices. Automated assistive tools and devices have shown to be extremely effective and necessary in the treatment of physical injuries and impairments. Physiotherapists might use android technology to remotely set the degrees of flexion/extension, the pace of movement, and duration. The Android application will also save the patient's exercise history and can convey the patient's input about his condition's progress to the physiotherapist. The normal active knee range of motion is as follows: Knee Flexion ROM: 135 deg (completely bent), Knee Extension ROM: 0 degrees (totally straight). External Knee Rotation ROM: 30-40 deg, Internal Knee Rotation ROM: 10deg Normal passive knee ROM is as follows: Passive Knee Flexion ROM: up to 150deg, depending on leg size - the limit is the calf pushing towards the back of the thigh, Passive knee extension range of motion (ROM): up to 10deg hyperextension is deemed typical. 30cm.

Place Hold on Design of Passive Controlled Iot Based Smart Cpm for Lower Limb /

Oral Administration of Fluoxetine Incorporated Liposomal Nanoparticles coated with PEG in Treatment of Chronic Mild stress (CMS) /

By Sadiq, Nadia . . 55p. , Depression is one of the most increasing psychological mood disorders, depression does not have some specific physical symptoms, but it negatively affects the person mood, way of living and thinking chronic mild stress (CMS) is one of its type, depression is affecting more than 300million people in the world. Shortness of breath, headache, being miserable, stomach disturbance and physical tensions are some of the common symptoms of depression that is reported according to National Alliance of Mental Health, Anti-depressant are used for altering the neurotransmitters those are serotonin, dopamine and norepinephrine that are primarily involved in regulating and mood alleviating mood. Despite the availability of large number of drugs, many of patients are resistant to the current mode of treatment that are available now adays. The obstacle that is significant for the transportation of beneficial therapeutic entities to the nervous system is BBB that is the blood brain barrier. There are junctions that are present in the endothelial cells of the blood brain barrier that stop the drugs passage. Now nanoparticles are receiving significant limelight owing to their small size and efficient brain targeting activity, making them likely to cross the blood brain barrier while carrying the intact drug molecule otherwise incapable of permeation. In this study the fluoxetine loaded liposomal nanoparticle coated with PEG were developed to transport the drug across the BBB to the central nervous system having much greater efficiency. For testing its delivery, the animal that is mice model of depression is used with the specific name that is designed in order to induce depression like symptoms in mice. Before and After treatment great differences between the physical behaviors like Elevated maze, open field test, force swim test etc and sucrose consumption test were identified and plotted, as liposomes that having capabilities to carry the anti-depressant drug molecule that is administered through the Oral route of administration or the better and improved method of treatment. 30cm.

Place Hold on Oral Administration of Fluoxetine Incorporated Liposomal Nanoparticles coated with PEG in Treatment of Chronic Mild stress (CMS) /

Analysis of post-COVID comorbidities in Pakistani population /

By Ansari, Sara. . 61p. , Introduction: Long-COVID, is the collective name given to denote persistence of symptoms for weeks or months in those who have recovered from SARS-COV-2 infection. If the relapsing symptoms are left undetected, can lead to chronic conditions. Still some work needs to be done to determine the correlation of the previous illnesses with the severity of post-COVID symptoms. Method: A questionnaire survey was spread amongst Pakistani population. About 83 COVID-19 survivors were included in the study, who were asked about any occurrence of symptoms at recovery, the experience and duration of the post-viral symptoms. Results: Out of 83 participants, 59(71.7%) experienced relapsing symptoms at 3 weeks from the onset of the viral infection, with 34(44.6%) having to face those symptoms for 3-4 weeks (postacute COVID). Females were significantly more likely to experience fatigue (p=0.014) and severity (p=0.032). The presence of symptoms was not associated with any therapy or activity. The presence of mild symptoms is common after the COVID-19 infection with those already suffering from anxiety, allergies, hypertension, and diabetes. Conclusion: This study highlights the importance of assessing those recovering from mild COVID-19 with acute-fatigue. Moreover, further longitudinal research in this area can help understand the management of chronic situations. 30cm.

Place Hold on Analysis of post-COVID comorbidities in Pakistani population /

Emg-Based Force Estimation Using Deep Learning Models /

By Nayab, Maham . . 94p. , The estimation of force through electromyography (EMG) assumes paramount importance in diverse domains, including neurorehabilitation, myoelectric control, and neurofeedback systems. The intricate relationship between muscle contraction and force, characterized by linear associations in small muscles with narrow motor units and nonlinear relationships in larger muscles with wider motor units, underscores the complexity of this physiological interplay. Against the backdrop of a global demand for advanced technologies to address limb loss limitations, with an estimated 100 million individuals worldwide in need of prosthetics, there arises an urgent need for sophisticated solutions. Meeting the diverse needs of prosthetic users emphasizes the crucial role of EMG-based force prediction, striving to provide adaptive and personalized solutions for an inclusive and effective approach to limb rehabilitation. This comprehensive study explores the dynamic interplay between surface electromyography (sEMG) and intramuscular electromyography (iEMG) signals and force estimation. Leveraging a diverse set of machine learning and deep learning models, the research aims to predict forces in both healthy individuals and those with trans-radial amputations. Across sEMG and iEMG modalities, deep learning models, including Long Short-Term Memory (LSTM), Temporal Convolutional Network (TCN), and the hybrid LSTM-TCN, consistently exhibit remarkable efficacy. These models, boasting R² values surpassing 0.90 in force prediction, offer promising advancements in refining force estimation through electromyography. Notably, the TCN emerges as an exemplary performer, yielding R² values of 0.98 for able-bodied individuals and 0.87 for amputees in sEMG. Simultaneously, the hybrid TCN-LSTM model maintains strong performance with R² values of 0.98 for able-bodied individuals and 0.85 for amputees in sEMG. The LSTM model also upholds notable performance, showcasing R² values of 0.99 for able-bodied individuals and 0.80 for amputees in sEMG. Beyond unraveling the intricacies of EMG-based force estimation, this study sheds light on the unique challenges posed by amputations, contributing substantively to the ongoing quest for enhanced precision and effectiveness in rehabilitation interventions. 30cm.

Place Hold on Emg-Based Force Estimation Using Deep Learning Models /

Classification of Activities of Daily Living (ADLs) Based Upper Limb Movements Using Machine Learning & Neural Network Classifiers /

By Anwer, Saba . . 68p. , Real time natural control of assistive, rehabilitation and prosthetic devices has gained significant importance over the last few decades for the people suffering from motor disabilities due to stroke, any spinal cord injury or amputation. Although surface electromyography (s-EMG) has been used as a viable controlling interface for several robotic devices specifically designed for post stroke therapeutic services. But these conventional controlling strategies are not feasible to design the rehabilitation or HMI systems based on simultaneous movements of multiple degrees of freedom (DOF). This paper presents a novel control strategy for HMIs which is based on the coupled use of EMG and inertial sensors. EMG and kinematic data of healthy and stroke subjects for commonly used daily life activities has been recorded. Multiple machine learning models including LDA, QDA, LSVM, QSVM, Fine KNN, Ensembled discriminant, and ensembled KNN have been applied. Besides this a tri-layered neural network classifier has also been implemented. A comparative analysis has been performed for the classification outcomes of all the applied models for EMG, IMU & EMG+IMU data. Overall, the KNN model performed well for all types of datasets with an average accuracy of 98.5% but results clearly demonstrated that average classification accuracies for all the applied models have significantly improved for EMG+IMU data which indicates that sensor fusion based control strategy for prosthesis can achieve higher performance than conventional control systems for each task. This study is an effort to provide a new EMG+IMU based technique for fast, efficient, and reliable control of robotic, rehabilitation and assistive devices for multiple movements with varying DOF. 30cm.

Place Hold on Classification of Activities of Daily Living (ADLs) Based Upper Limb Movements Using Machine Learning & Neural Network Classifiers /

Classification of Reach and Grasp Motions from EEG Signals using Deep Convolutional Neural Networks (CNN) /

By Sultan, Hajrah . . 73p. , Classification of neural correlates of hand motions from EEG signals recently gained attention of researchers for the development of BCI systems for persons suffered from stroke, spinal cord injury who are not able to do voluntary movements or person with amputated arm or legs. Commonly LDA, SVM and K-NN models are used for the classification of hand motions, CNN and hybrid models are also used but most methods include the complex methods or preprocessing of EEG data and extraction of time or frequency domain features from the preprocessed signals which is a time consuming and lack flexibility because the EEG signals vary from human to human. In this thesis a Deep CNN model for end-to-end0learning of neural corelates for reach and grasp actions is introduced, aiming to increase rate of recognition & balanced classification0accuracy throughout all the subjects. A new model of CNN for movement classification is proposed that can also be used on the edge devices because of its smaller size for the development of BCI systems. In the proposed model separable convolutional blocks are used which reduce the number of parameters and hence the size of model also decreases. The dataset that is used for the testing of model is BNCI Horizon 2020 Reach and Grasp action dataset that is publicly available dataset. The dataset is also tested on 3 machine learning models LDA, SVM and K-NN are used, in which input is given in the form of time domain feature set the average accuracies achieved on these models are 60.77 (±3.80 STD), 66.73 (±2.86 STD), and 79.81 (±3.11 STD) respectively on the unseen dataset. Then the dataset is tasted on proposed Deep learning model along with DeepConvNet and EffNet models. The proposed model achieves the average classification accuracy of 92.44 (±4.13 STD), 92.9 (±4.23 STD) and 81.7(±5.68 STD). The model proposed achieves the same accuracy as DeepConvNet, but the size of proposed model is far smaller than the DeepConvNet model. Results shows that the proposed model shows the improved results with less variation of results within the subjects. Which will become helpful in the creation of real time BCI systems. 30cm.

Place Hold on Classification of Reach and Grasp Motions from EEG Signals using Deep Convolutional Neural Networks (CNN) /

Automated Brain Tumor Segmentation using Multimodal MRI Scans /

By Ehsan ,Fatima . . 73p. ; 30cm..

Place Hold on Automated Brain Tumor Segmentation using Multimodal MRI Scans /

Hemodynamic Signal Prediction Using Modified General Linear Model /

By SABA ,NAVEED E. . 75p. ; 33cm..

Place Hold on Hemodynamic Signal Prediction Using Modified General Linear Model /

DESIGN AND DEVELOPMENT OF SOFT EXOSKELETAL GLOVE /

By AKHTER ,MALIK NAVEED. . 43p. ; 33cm..

Place Hold on DESIGN AND DEVELOPMENT OF SOFT EXOSKELETAL GLOVE /

Design Development and Fabrication of Bileaflet Mechanical Mitral Heart Valve /

By Natasha . . 65p. ; 33cm..

Place Hold on Design Development and Fabrication of Bileaflet Mechanical Mitral Heart Valve /

AUTOMATIC DETECTION OF PATHOLOGICAL MYOPIA USING DEEP LEARNING /

By RAUF ,NAMRA . . 55p. ; 33cm..

Place Hold on AUTOMATIC DETECTION OF PATHOLOGICAL MYOPIA USING DEEP LEARNING /

DEVELOPMENT OF IMPEDANCE METER FOR BIOMEDICAL APPALICATIONS /

By RIZVI ,S.QAISAR ABBAS. . 51p. ; Soft Copy , .

Place Hold on DEVELOPMENT OF IMPEDANCE METER FOR BIOMEDICAL APPALICATIONS /

Functional sensitivity of Multiband Echo-planar imaging in subcortical region /

By Ullah, Azmat . . 62p. ; 30cm..

Place Hold on Functional sensitivity of Multiband Echo-planar imaging in subcortical region /

Finite Element Modelling of Foot Bones In A Patient With Chronic Osteomyelitis /

By JAHANGIR ,SANA . . 104p. ; 30cm..

Place Hold on Finite Element Modelling of Foot Bones In A Patient With Chronic Osteomyelitis /

FNIRS DATA CLASSIFICATION FOR BRAIN COMPUTER INTERFACE USING DEEP LEARNING /

By SUBHANI , AHMAD . . 67p. ; 30cm..

Place Hold on FNIRS DATA CLASSIFICATION FOR BRAIN COMPUTER INTERFACE USING DEEP LEARNING /

Design and Manufacturing of Ankle Rehabilitation Device for Ankle Support /

By ASAD, MALIHA . . 51p. ; 30cm..

Place Hold on Design and Manufacturing of Ankle Rehabilitation Device for Ankle Support /

Decoding of Hand Motion Using State of Art Time Domain, Frequency Domain And Feature Extraction /

By Munawar, Sulaiman . . 69p. , Exoskeletons that are activated by the muscles and brain have been suggested to train the motor skills of stroke victims. Training can incorporate task variety since an exoskeleton allows for the execution of various movement types.Differentiating between movement types at the same time from brain activity is challenging, but it might be accessible from residual muscular activity that many patients retain regain.This study examines whether forearm EMG from five stroke patients can be used to decode seven distinct motion classes of the hand and forearm. This study evaluates classifiers like Support vector machine (SVM), Lineardiscriminant analysis (LDA) and K nearest neighbor (KNN). It investigated the relation of motor impairment with classification accuracy by the classifiers. During the following motion classes: Supination, Pronation, Hand Close, Hand Open, Wrist Extension, Wrist Flexion, and Pich, five surface EMG channels were recorded.Every motion was performed by patients three times repetition over the course of eight weeks.Support vector machines, k nearest neighbor, and linear discriminant analysis were used to classify decoding of hand moments for stroke patients. On average,73.69 ± 6.39%SVM,71.6 ± 5.09% KNNand 50±4.56 LDA of the movements were correctly classified.Seven motion classes were demonstrated to be decoded from residual EMG, and SVM proved to be the most effective classification method when compared to the other three classifiers for decoding of hand motion for stroke patients.The results of this study may have implications for the development of exoskeletons, suits, or gadgets, that are powered by EMG signals. These devices might be utilized in the comfort of the patient's home to assist stroke sufferers with their training activities. Therefore, the findings of this study may assist in improving the effectiveness and accessibility of these useful tools for stroke survivors. 30cm.

Place Hold on Decoding of Hand Motion Using State of Art Time Domain, Frequency Domain And Feature Extraction /

Dexamethasone on loaded transfersomes nanoparticles for skin delivery /

By Azam, Nayab. . 79p. ; 30cm..

Place Hold on Dexamethasone on loaded transfersomes nanoparticles for skin delivery /

Predicting Risk of Perinatal Depression and Anxiety Using Biomedical, Psychosocial & Socioeconomic Factors Through Machine Learning /

By JAVED ,FAJAR. . 72p. ; 30cm..

Place Hold on Predicting Risk of Perinatal Depression and Anxiety Using Biomedical, Psychosocial & Socioeconomic Factors Through Machine Learning /

Discovery of Potent LasR Inhibitor Using Pharmacophore Informatics Strategy /

By Sadiq , Suhaib. . 103p. ; 30cm..

Place Hold on Discovery of Potent LasR Inhibitor Using Pharmacophore Informatics Strategy /

Computational Modeling of Human Head Model for Correlation Study of Cerebrospinal Fluid and Tumor /

By Uzair Ul Haq ,Muhammad . . 68p. ; 30cm..

Place Hold on Computational Modeling of Human Head Model for Correlation Study of Cerebrospinal Fluid and Tumor /

Finite Element Analysis of Intracranial Pressure under the influence of Brain Tumor /

By Ahmed ,Ali . . 87p. ; 30cm..

Place Hold on Finite Element Analysis of Intracranial Pressure under the influence of Brain Tumor /

CALCULATION OF FRACTIONAL FLOW RESERVE (FFR) USING CT ANGIOGRAPHY IMAGES /

By WAQAR ,SHEIKH ARSLAN. . 59p. ; 30cm..

Place Hold on CALCULATION OF FRACTIONAL FLOW RESERVE (FFR) USING CT ANGIOGRAPHY IMAGES /

Lung Cancer Detection Using Digital Image Processing on CT Soan Image /

By Shoaib ,Muhammad. . 40p. ; 30cm..

Place Hold on Lung Cancer Detection Using Digital Image Processing on CT Soan Image /

Decoding Lower Limb Residual Muscle Activity in Chronic Stroke /

By Noor ,Afaq Ahmed. . 50P. ; 30cm..

Place Hold on Decoding Lower Limb Residual Muscle Activity in Chronic Stroke /

Real-time approach for the effect of arm position using Fitts’ Law /

By Ahmad , Jamil . . 50p. ; 30cm..

Place Hold on Real-time approach for the effect of arm position using Fitts’ Law /

Study of various gas concentrations in exhaled breath of hemodialysis patients /

By Ahmad ,Osama . . 79p. ; 30cm..

Place Hold on Study of various gas concentrations in exhaled breath of hemodialysis patients /

Investigation of Ablation Techniques for Different Types of Brain Tumors /

By BINTE IRFAN ,MANAHIL . . 41p. ; 30cm..

Place Hold on Investigation of Ablation Techniques for Different Types of Brain Tumors /

Melanoma Detection Using Machine Learning /

By FAROOQ SAEED ,MUHAMMAD . . 45p. ; 30cm..

Place Hold on Melanoma Detection Using Machine Learning /

Machine/Deep Learning Based Adaptive Myoelectric Control /

By Ashraf ,Hassan . . 56p. ; 30cm..

Place Hold on Machine/Deep Learning Based Adaptive Myoelectric Control /

Design and Development of Portable Incubator /

By Ilyas ,Yasir. . 68p. ; 30cm..

Place Hold on Design and Development of Portable Incubator /

Modeling and Simulation of Rehabilitation/Assistive Device for Stroke Effected Hand /

By Faiz ,Muhammad Sohaib. . 47p. ; 30cm..

Place Hold on Modeling and Simulation of Rehabilitation/Assistive Device for Stroke Effected Hand /

Detection of Local Muscle Fatigue in Writing Task Using Surface Electromyography /

By Abbas ,Nadir . . 46p. ; 30cm..

Place Hold on Detection of Local Muscle Fatigue in Writing Task Using Surface Electromyography /

Development of Portable Photo Spectrometer for Classification of Biomedical Fluids using Neural Network /

By Rasool ,Hassan Nawazish. . 67p. ; 30cm..

Place Hold on Development of Portable Photo Spectrometer for Classification of Biomedical Fluids using Neural Network /

Vein Sight – Infrared based vein imaging system /

By Jan ,Khizer . . 48p. ; 30cm..

Place Hold on Vein Sight – Infrared based vein imaging system /

A Comparative Study of Asthmatic and Normal Subjects for Feasible Device Development to Predict Asthma Attack /

By Sarwar, Samrina . . 47p. ; 30cm..

Place Hold on A Comparative Study of Asthmatic and Normal Subjects for Feasible Device Development to Predict Asthma Attack /

Multi-Disease Classification For Retinal Diseases Using Deep Learning Technique

By Khan, Omar Salman . . 55p. , Diagnosis before the spread of retinal diseases is vital to prevent high level blindness and any sort of visual impairment. Many retinal diseases can be found via the fundus imaging which has a very important role in the observation and detection of various ophthalmic diseases. Most previous literature has focused their approaches on identifying individual diseases or a combination of 3-4 diseases like DR, MYA, ARMD, MH, ODC having major research. The eye is mostly affected by more than one underlying disease or disease marker, and uptil now most datasets had very few classes. Recently introduced RFMiD dataset, is one of the first datasets to provide 45 different classes of ophthalmic diseases. Hence making it possible to work towards automated multi-disease classification models which would provide great help to highlight this issue via clinical decision support systems integrated in the medical image diagnosis. Our work aimed to achieve higher accuracy than previous literature and to create an CDS application from the model in understanding and predicting multi retinal diseases. Deep learning models are excellent and have proven to be extremely effective in solving complex image processing problems. In addition, ensemble learning yields high generalization performance by reducing variance. Therefore, a synthesis of transfer, ensemble, and deep learning was used in this work to create an accurate and reliable model for multi retinal disease classification. To create the Multi Retinal Disease Classification Model (MRDCM) we used ensemble of EfficientNetB4 and EfficientNetV2S, with our final ensemble model giving promising results. In our evaluation, we scored an AUC of 0.973 which stands better than literature. Further our model selection is lighter than models used in literature. The model was tested on 27 main classes of RFMiD dataset for comparison with literature. Index Terms—Deep iii Learning, Ensemble learning, Retinal Image Analysis, multi-Disease classification, transfer learning. 30cm.

Place Hold on Multi-Disease Classification For Retinal Diseases Using Deep Learning Technique

Ocular Disease Intelligent Recognition /

By Sahar, Syeda Ghina . . 46p. , To record anatomical details of the eye and anomalies, fundus imaging has proved very efficient. The most effective way to see and diagnose a wide range of eye diseases is through fundus imaging. Conditions that affect the blood vessels and areas surrounding it include diabetesrelated retinopathy, glaucoma, AMD, myopia, cataract and hypertension. It's possible for the patient to have more than one ophthalmological problems that can be seen in one or both of his eyes. The dataset provided by ODIR is used in this study. The data has eight different categories for the diseases to be detected. By using transfer learning, two simultaneous models are described for solving the multi label problem for both the eyes (left and right). For the convolutional network, two synchronous efficient net models are implemented which are used with ADAM optimizers for better detection and results outcome. On the ODIR data set, B7 Efficient net along with focal loss outperformed the other approaches with an accuracy rate of 0.96%. 30cm.

Place Hold on Ocular Disease Intelligent Recognition /

3D Neural Network for Detection of ACL Injury in Knee MRI Scans /

By Kamran, Abdullah . . 43p. , Computer aided diagnosis is widely used in medical imaging for the diagnosis of many diseases such as cardiomegaly, brain and kidney tumor, lung cancer, COVID-19 and may more. For the past few decades, computer aided diagnosis has significantly improved due to the development of better architecture used for the diagnosis. Knee injury diagnosis using deep learning techniques is highly popular due its high detection rate and is highly localized. Many state-of-the-art-deep learning models have been used for the detection of abnormalities, meniscus tear and ACL tears in Knee MRI scans. These models include RESNET, Google-Net, VGG19 and VGG16, Alex-Net and many other, all giving significant results. In this study we used a custom 3D CNN model which is light in weight. For training we are using two datasets, one provided by Stanford ML group and the other form Hospital in Croatia. We combined the two dataset and split it into 80-20 ration (80% of the data used for training and remaining for testing purposes). Both the dataset has extreme class imbalance, so we used data augmentation and class weights to rectify its effect on the training process. Further the voxel intensities for the two datasets were different (one dataset was in 8-bit format and the second was in 12-bit format), so we normalized the intensity values using mathematical formulas. For contrast, we performed adaptive histogram equalization Average accuracy and AUC achieved by our model on training set is 97.6 and 99.3 respectively, during 5-fold cross validation. 30cm.

Place Hold on 3D Neural Network for Detection of ACL Injury in Knee MRI Scans /