Mohamed Abdelmoniem

Fahmida Haque, Mamun Bin Ibne Reaz, Muhammad Enamul Hoque Chowdhury et al.

Diabetic sensorimotor polyneuropathy (DSPN) is one of the prevalent forms of neuropathy affected by diabetic patients that involves alterations in biomechanical changes in human gait. In literature, for the last 50 years, researchers are trying to observe the biomechanical changes due to DSPN by studying muscle electromyography (EMG), and ground reaction forces (GRF). However, the literature is contradictory. In such a scenario, we are proposing to use Machine learning techniques to identify DSPN patients by using EMG, and GRF data. We have collected a dataset consists of three lower limb muscles EMG (tibialis anterior (TA), vastus lateralis (VL), gastrocnemius medialis (GM) and 3-dimensional GRF components (GRFx, GRFy, and GRFz). Raw EMG and GRF signals were preprocessed, and a newly proposed feature extraction technique scheme from literature was applied to extract the best features from the signals. The extracted feature list was ranked using Relief feature ranking techniques, and highly correlated features were removed. We have trained different ML models to find out the best-performing model and optimized that model. We trained the optimized ML models for different combinations of muscles and GRF components features, and the performance matrix was evaluated. This study has found ensemble classifier model was performing in identifying DSPN Severity, and we optimized it before training. For EMG analysis, we have found the best accuracy of 92.89% using the Top 14 features for features from GL, VL and TA muscles combined. In the GRF analysis, the model showed 94.78% accuracy by using the Top 15 features for the feature combinations extracted from GRFx, GRFy and GRFz signals. The performance of ML-based DSPN severity classification models, improved significantly, indicating their reliability in DSPN severity classification, for biomechanical data.

Tariq O. Abbas, Mohamed AbdelMoniem, Ibrahim Khalil et al.

Objectives: To explore the capacity of deep learning algorithm to further streamline and optimize urethral plate (UP) quality appraisal on 2D images using the plate objective scoring tool (POST), aiming to increase the objectivity and reproducibility of UP appraisal in hypospadias repair. Methods: The five key POST landmarks were marked by specialists in a 691-image dataset of prepubertal boys undergoing primary hypospadias repair. This dataset was then used to develop and validate a deep learning-based landmark detection model. The proposed framework begins with glans localization and detection, where the input image is cropped using the predicted bounding box. Next, a deep convolutional neural network (CNN) architecture is used to predict the coordinates of the five POST landmarks. These predicted landmarks are then used to assess UP quality in distal hypospadias. Results: The proposed model accurately localized the glans area, with a mean average precision (mAP) of 99.5% and an overall sensitivity of 99.1%. A normalized mean error (NME) of 0.07152 was achieved in predicting the coordinates of the landmarks, with a mean squared error (MSE) of 0.001 and a 20.2% failure rate at a threshold of 0.1 NME. Conclusions: This deep learning application shows robustness and high precision in using POST to appraise UP quality. Further assessment using international multi-centre image-based databases is ongoing. External validation could benefit deep learning algorithms and lead to better assessments, decision-making and predictions for surgical outcomes.

Fahd Seddik, Abdulrahman Elbedewy, Gaser Sami et al.

Training modern deep learning models is increasingly constrained by GPU memory and compute limits. While Randomized Numerical Linear Algebra (RandNLA) offers proven techniques to compress these models, the lack of a unified, production-grade library prevents widely adopting these methods. We present Panther, a PyTorch-compatible library that consolidates established RandNLA algorithms into a single high-performance framework. Panther engineers efficient, drop-in replacements for standard components including sketched linear layers, 2D convolution, multi-head attention, and randomized matrix decompositions (such as pivoted CholeskyQR). By implementing a custom C++/CUDA backend (pawX), Panther provides an optimized implementation that can run on both CPUs and GPUs. We demonstrate the effectiveness of RandNLA techniques and Panther's ease of adoption. By replacing standard PyTorch linear layers with Panther layers (requiring only a few lines of code) we achieve significant memory savings (up to 75%) on BERT while maintaining comparable loss. Source code is available (MIT License) at https://github.com/FahdSeddik/panther, along with demonstration video at https://youtu.be/7M3RQb4KWxs.