Showing 1–3 of 3 results for author: Chakma, A

Benchmarking Deep Learning Frameworks for Automated Diagnosis of Ocular Toxoplasmosis: A Comprehensive Approach to Classification and Segmentation

Authors: Syed Samiul Alam, Samiul Based Shuvo, Shams Nafisa Ali, Fardeen Ahmed, Arbil Chakma, Yeong Min Jang

Abstract: Ocular Toxoplasmosis (OT), is a common eye infection caused by T. gondii that can cause vision problems. Diagnosis is typically done through a clinical examination and imaging, but these methods can be complicated and costly, requiring trained personnel. To address this issue, we have created a benchmark study that evaluates the effectiveness of existing pre-trained networks using transfer learnin… ▽ More Ocular Toxoplasmosis (OT), is a common eye infection caused by T. gondii that can cause vision problems. Diagnosis is typically done through a clinical examination and imaging, but these methods can be complicated and costly, requiring trained personnel. To address this issue, we have created a benchmark study that evaluates the effectiveness of existing pre-trained networks using transfer learning techniques to detect OT from fundus images. Furthermore, we have also analysed the performance of transfer-learning based segmentation networks to segment lesions in the images. This research seeks to provide a guide for future researchers looking to utilise DL techniques and develop a cheap, automated, easy-to-use, and accurate diagnostic method. We have performed in-depth analysis of different feature extraction techniques in order to find the most optimal one for OT classification and segmentation of lesions. For classification tasks, we have evaluated pre-trained models such as VGG16, MobileNetV2, InceptionV3, ResNet50, and DenseNet121 models. Among them, MobileNetV2 outperformed all other models in terms of Accuracy (Acc), Recall, and F1 Score outperforming the second-best model, InceptionV3 by 0.7% higher Acc. However, DenseNet121 achieved the best result in terms of Precision, which was 0.1% higher than MobileNetv2. For the segmentation task, this work has exploited U-Net architecture. In order to utilize transfer learning the encoder block of the traditional U-Net was replaced by MobileNetV2, InceptionV3, ResNet34, and VGG16 to evaluate different architectures moreover two different two different loss functions (Dice loss and Jaccard loss) were exploited in order to find the most optimal one. The MobileNetV2/U-Net outperformed ResNet34 by 0.5% and 2.1% in terms of Acc and Dice Score, respectively when Jaccard loss function is employed during the training. △ Less

Submitted 18 May, 2023; originally announced May 2023.

NRC-Net: Automated noise robust cardio net for detecting valvular cardiac diseases using optimum transformation method with heart sound signals

Authors: Samiul Based Shuvo, Syed Samiul Alam, Syeda Umme Ayman, Arbil Chakma, Prabal Datta Barua, U Rajendra Acharya

Abstract: Cardiovascular diseases (CVDs) can be effectively treated when detected early, reducing mortality rates significantly. Traditionally, phonocardiogram (PCG) signals have been utilized for detecting cardiovascular disease due to their cost-effectiveness and simplicity. Nevertheless, various environmental and physiological noises frequently affect the PCG signals, compromising their essential distinc… ▽ More Cardiovascular diseases (CVDs) can be effectively treated when detected early, reducing mortality rates significantly. Traditionally, phonocardiogram (PCG) signals have been utilized for detecting cardiovascular disease due to their cost-effectiveness and simplicity. Nevertheless, various environmental and physiological noises frequently affect the PCG signals, compromising their essential distinctive characteristics. The prevalence of this issue in overcrowded and resource-constrained hospitals can compromise the accuracy of medical diagnoses. Therefore, this study aims to discover the optimal transformation method for detecting CVDs using noisy heart sound signals and propose a noise robust network to improve the CVDs classification performance.For the identification of the optimal transformation method for noisy heart sound data mel-frequency cepstral coefficients (MFCCs), short-time Fourier transform (STFT), constant-Q nonstationary Gabor transform (CQT) and continuous wavelet transform (CWT) has been used with VGG16. Furthermore, we propose a novel convolutional recurrent neural network (CRNN) architecture called noise robust cardio net (NRC-Net), which is a lightweight model to classify mitral regurgitation, aortic stenosis, mitral stenosis, mitral valve prolapse, and normal heart sounds using PCG signals contaminated with respiratory and random noises. An attention block is included to extract important temporal and spatial features from the noisy corrupted heart sound.The results of this study indicate that,CWT is the optimal transformation method for noisy heart sound signals. When evaluated on the GitHub heart sound dataset, CWT demonstrates an accuracy of 95.69% for VGG16, which is 1.95% better than the second-best CQT transformation technique. Moreover, our proposed NRC-Net with CWT obtained an accuracy of 97.4%, which is 1.71% higher than the VGG16. △ Less

Submitted 28 April, 2023; originally announced May 2023.

Domain Adaptation for Inertial Measurement Unit-based Human Activity Recognition: A Survey

Authors: Avijoy Chakma, Abu Zaher Md Faridee, Indrajeet Ghosh, Nirmalya Roy

Abstract: Machine learning-based wearable human activity recognition (WHAR) models enable the development of various smart and connected community applications such as sleep pattern monitoring, medication reminders, cognitive health assessment, sports analytics, etc. However, the widespread adoption of these WHAR models is impeded by their degraded performance in the presence of data distribution heterogene… ▽ More Machine learning-based wearable human activity recognition (WHAR) models enable the development of various smart and connected community applications such as sleep pattern monitoring, medication reminders, cognitive health assessment, sports analytics, etc. However, the widespread adoption of these WHAR models is impeded by their degraded performance in the presence of data distribution heterogeneities caused by the sensor placement at different body positions, inherent biases and heterogeneities across devices, and personal and environmental diversities. Various traditional machine learning algorithms and transfer learning techniques have been proposed in the literature to address the underpinning challenges of handling such data heterogeneities. Domain adaptation is one such transfer learning techniques that has gained significant popularity in recent literature. In this paper, we survey the recent progress of domain adaptation techniques in the Inertial Measurement Unit (IMU)-based human activity recognition area, discuss potential future directions. △ Less

Submitted 6 April, 2023; originally announced April 2023.