Student Research Symposium    

At the 20th Annual Student Research Symposium organized by Texas A&M University at Galveston, three CARES Lab students, supervised by Dr. Irfan Khan in MARE 491: Research and collaborating with graduate students Bilal Zahid Hussain and Shorya Agarwal; were selected for the Computer Science Award.

Mathew Vincent examined interpretable neural network architectures paired with SHapley Additive exPlanations to extract clinical insights from Behavioral and Functional Respiratory Symptom Survey data. He conducted a comparative analysis against an XGBoost baseline, evaluating area under the receiver operating characteristic curve and precision-recall metrics to assess transparency alongside prediction performance.

Aman Sriven developed a hybrid residual-attention ensemble framework for smart contract security. He fed Ethereum bytecode sequences into convolutional residual blocks and self-attention modules. Ensemble learners integrated transformer-style layers to detect vulnerabilities such as reentrancy, integer overflow and unchecked calls. He reported precision, recall and F1-score through k-fold cross-validation.

Aditya Seelaboyina applied an ensemble of pretrained convolutional neural networks fine-tuned on the NIH Chest X-ray dataset. He combined ResNet50, DenseNet121 and EfficientNet-B0 models, weighted their softmax outputs for 14 chest disease labels. He employed data augmentation: random rotations, flips and contrast adjustments; to improve generalization. He evaluated results using area under the precision-recall curve and class-wise sensitivity.

All three students delivered outstanding presentations, showcasing rigorous methodology, clear insights and strong performance metrics that reflected the innovation and quality of CARES Lab’s research.