Deepak Mishra

This letter presents a novel, integrated multimodal data acquisition system for simultaneously capturing the six degrees of freedom motion data and real-time visual information, designed primarily for ultrasound imaging applications. The system combines inertial measurement units, ArUco markers, and a video capture device to achieve high-accuracy motion tracking synchronized with real-time ultrasound imaging. Our approach provides a cost-effective and portable setup that is capable of recording accurate translational and rotational data. The experimental results demonstrate promising results with $1.95\pm \text{1.10}\,\text{mm}$ deviation with the path driven by the robotic manipulation, which can be further improved by controlling acceleration and velocity. Its real-time performance, ease of use, and potential for AI model training make it valuable for various medical applications, including ultrasound-guided procedures and motion analysis.

CNNs have become the standard for medical image interpretation, but concerns persist about their reliability in real-world applications. CNNs can be sensitive to small variations in image quality and vulnerable to adversarial attacks, potentially leading to inaccurate diagnoses. To address these issues, we introduce a novel scale-aware adaptive feature quantization approach. This enhances the robustness and reliability of CNNs by adaptively combining quantized representations from multiple scales, improving performance on low-quality or perturbed images. Our approach uses soft codes and dynamic weighting to adaptively combine features from different scales, creating a more informative final quantized representation. Experimental results on diverse medical datasets, including chest X-rays and dermatoscopic images, demonstrate the effectiveness of our approach. Our method significantly outperforms both standard CNNs and state-of-the-art approaches, with substantial gains across all metrics (AUC, F1 score). These improvements range from 2.6% to 11%, demonstrating our method’s superior performance and reliability for medical diagnosis in challenging real-world scenarios.

Knowledge distillation is a common technique for improving the performance of a shallow student network by transferring information from a teacher network, which in general, is comparatively large and deep. These teacher networks are pre-trained and often uncalibrated, as no calibration technique is applied to the teacher model while training. Calibration of a network measures the probability of correctness for any of its predictions, which is crucial for high-risk domains where reliable predictions are essential. In this paper, we study how to obtain a calibrated student from an uncalibrated teacher. Our approach relies on the fusion of the data-augmentation techniques, including but not limited to Mixup and CutMix, with knowledge distillation. We incorporate and analyze the impact of focal loss in the distillation framework to further improve the calibration of the student model. We perform extensive experiments to validate our approach on various datasets, including CIFAR-100, TinyImageNet, ImageNet and Diabetic Retinopathy datasets, and compare it with various techniques like CRD, RKD, DKD and MLLD to obtain calibrated student models. Furthermore, we conduct an ablation study to dissect the influence of augmentation techniques and the integration of focal loss. Additionally, we assess the robustness of our approach by evaluating its performance on corrupted CIFAR-100C data, demonstrating its consistent and reliable outcomes even under challenging conditions.

Federated Learning (FL) enables collaborative model training across decentralized data sources while preserving privacy. However, FL systems are vulnerable to attacks from malicious clients that can degrade model performance and compromise integrity. In this work, we propose AR2FL, an anomaly-resistant and robust framework that enhances FL aggregation by leveraging mean latent representations of client updates. This data-driven approach enables the server to estimate inter-client similarity and dynamically scale clients contributions, reducing the influence of anomalous or adversarial updates. Unlike methods based on fixed distance metrics such as cosine similarity or Euclidean distance, AR2FL captures deeper statistical patterns in the latent space, enabling more accurate and secure model updates. Experiments on several datasets show AR2FL maintains strong accuracy, fast convergence, and high robustness, making it suitable for secure large-scale FL.

Self-supervised learning (SSL) has gained traction in medical image analysis, enabling representation learning with limited labels. While contrastive SSL, using diverse augmentations, has become dominant, we argue that applying standard augmentations, originally designed for natural images, to medical images like chest X-rays is suboptimal. Chest X-rays possess unique structures and subtle abnormalities that differ from natural images, and preserving these during augmentation is critical for learning clinically meaningful representations. In this paper, we introduce a novel set of domain-specific contextual transformations tailored for chest X-rays, including anatomy-aware perturbations and context random masking, designed to preserve diagnostic semantics during SSL pre-training. This augmentation strategy is the first to explicitly align transformation design with radiological context, addressing a major gap in existing medical SSL approaches. Experiments on NIH, RSNA, and SIIM datasets show that our approach yields up to 5% improvement in downstream tasks under limited supervision, compared to standard augmentations.

Federated Learning (FL) has emerged as a promising paradigm for collaborative and privacy-preserving model training in medical imaging. However, FL faces major challenges such as data heterogeneity among hospitals or institutions and scarcity of labeled data, particularly in healthcare applications. To address these challenges, we propose FLAME (Federated Learning with masked Autoencoders and Mean-prototypes Embedding) for sparsely labeled medical images. FLAME implements an integrated learning framework where a masked autoencoder (MAE) learns robust feature representations through reconstruction-based self-supervision, while a Prototypical Network head guides these representations to enhance class separation through mean-prototype embeddings. This learning mechanism enables the encoder to simultaneously capture rich contextual features from unlabeled data while learning discriminative boundaries among classes using limited labeled samples. Our experiments on diverse medical imaging tasks, including PathMNIST, Dermnet, COVID-19 chest X-ray dataset, and Skin-FL, demonstrate FLAME's superior performance over existing FL techniques. The framework shows significant improvements in both classification accuracy and convergence speed, while maintaining privacy and reducing dependence on labeled data. Most importantly, the proposed integration of MAE and Prototypical Network opens new possibilities for the domains that suffer from label scarcity and data heterogeneity, making it particularly valuable for applications like medical diagnostics.