Personal protective equipment (PPE) is critical for infection control in healthcare, protecting workers and patients from infection risks. The COVID-19 pandemic further highlighted the importance of correct PPE use, yet adherence to U.S. Centers for Disease Control and Prevention guidelines remains inconsistent. Continuous human monitoring of PPE adherence is impractical because it is labor-intensive and may expose observers to infection risk. Automated monitoring is a promising alternative, but reliable PPE assessment in clinical videos remains difficult due to occlusion and subtle differences between adherence levels. To address these challenges, we propose SAFE - Smart Adherence detection Framework for PPE, a cascaded computer vision system for real-time monitoring of PPE wearing status, including complete, incomplete, and absent cases, with a focus on gowns and masks. SAFE uses a two-stage design: Stage 1 detects gown status and localizes head regions, and Stage 2 classifies mask status from head crops. We evaluate SAFE on R2PPE, a ceiling-view trauma-room simulation dataset with dense PPE annotations and complex scenes. SAFE improves overall average precision from 0.48 to 0.67 and increases mask-class average precision by 0.33 compared to a baseline one-stage detector. We further validate SAFE across modern detector backbones, including transformer-based detectors, and on real-case trauma-room data using class-level and alarm-level criteria, improving class-level mask accuracy from 0.59 to 0.65 while maintaining a high alarm-level recall of 0.98. SAFE could enhance PPE monitoring with minimal human intervention, providing a scalable solution for improving infection control in healthcare settings.

Facial expression recognition (FER) models often face challenges when generalizing across domains, such as different datasets and age groups. Despite the significance of this problem, FER in children (child FER) research remains relatively understudied, and such studies exhibit vulnerability to cross-domain evaluation. In response to the scarcity of child FER research, we propose a novel domain-agnostic approach for robust child FER. The architecture integrates a child-centric source-domain reconstructor and a child emotion feature-guided classifier. First, we use a secondary image diffusion model to reconstruct the image with source-domain childlike features while preserving emotion. Second, we recognize facial expressions based on domain-agnostic features from reconstructed images through a cross-attention mechanism. This approach counters performance degradation caused by domain discrepancies and improves the generalizability of child FER. We validate the proposed approach with diverse, publicly available datasets to highlight its effectiveness. The source code is available at https://github.com/st0421/Child-FER.

Chest X-rays (CXRs) are widely used imaging modality for the diagnosis and prognosis of lung disease. There is a large body of work where machine learning algorithms are developed for specific tasks. However, the traditional diagnostic tool design methods based on supervised learning are burdened by the need to provide training data annotation, which should be of good quality for better clinical outcomes. Here, we propose an alternative solution, a new self-supervised paradigm, where a general representation from CXRs is learned using a group-masked self-supervised framework. The pre-trained model is then fine-tuned for domain-specific tasks such as covid-19, pneumonia detection, and general health screening. We show that the same pre-training can be used for the lung segmentation task. Our proposed paradigm shows robust performance in multiple downstream tasks which demonstrates the success of the pre-training. Moreover, the performance of the pre-trained models on data with significant drift during test time proves the learning of a better generic representation. The methods are further validated by covid-19 detection in a unique small-scale pediatric data set. The performance gain ($\sim 25 \%$) is significant when compared to a supervised transformer-based method. This adds credence to the strength and reliability of our proposed framework and pre-training strategy.

Segmentation of a brain tumor from magnetic resonance multimodal images is a challenging task in the field of medical imaging. The vast diversity in potential target regions, appearance and multifarious intensity threshold levels of various tumor types are few of the major factors that affect segmentation results. An accurate diagnosis and its treatment demand strict delineation of the tumor affected tissues. Herein, we focus on a smart, automated, and robust segmentation approach for brain tumor using a modified 3D U-Net architecture. The pre-operative multimodal 3D-MRI scans of High-Grade Glioma (HGG) and Low-Grade Glioma (LGG) are used as data. Our proposed approach solves the problem of memory and system resource constraints by robustly applying dense network training on image patches of 3D volumes. It improves the border region artifact detection by applying convolutions at an appropriate phase in the proposed neural network. Multi-class imbalance data are handled by using Categorical Cross Entropy (CCE) loss developed by combining the Weighted Cross Entropy (WCE) with Weighted Multi-class Dice Loss (WMDL) functions, which enables the network to perform smart segmentation of the smaller tumorous regions. The proposed approach is tested and evaluated for the challenge datasets of multimodal MRI volumes of tumor patients. Experiments are performed to compute the average dice scores on BraTS-2019 and BraTS-2020 datasets for the whole tumor region.

We propose a novel capsule network based variational encoder architecture, called Bayesian capsules (B-Caps), to modulate the mean and standard deviation of the sampling distribution in the latent space. We hypothesized that this approach can learn a better representation of features in the latent space than traditional approaches. Our hypothesis was tested by using the learned latent variables for image reconstruction task, where for MNIST and Fashion-MNIST datasets, different classes were separated successfully in the latent space using our proposed model. Our experimental results have shown improved reconstruction and classification performances for both datasets adding credence to our hypothesis. We also showed that by increasing the latent space dimension, the proposed B-Caps was able to learn a better representation when compared to the traditional variational auto-encoders (VAE). Hence our results indicate the strength of capsule networks in representation learning which has never been examined under the VAE settings before.

Human stress is a serious health concern, which must be addressed with appropriate actions for a healthy society. This paper presents an experimental study to ascertain the appropriate phase, when electroencephalography (EEG) based data should be recorded for classification of perceived mental stress. The process involves data acquisition, pre-processing, feature extraction and selection, and classification. The stress level of each subject is recorded by using a standard perceived stress scale questionnaire, which is then used to label the EEG data. The data are divided into two (stressed and non-stressed) and three (non-stressed, mildly stressed, and stressed) classes. The EEG data of 28 participants are recorded using a commercially available four channel Muse EEG headband in two phases i.e., pre-activity and post-activity. Five feature groups, which include power spectral density, correlation, differential asymmetry, rational asymmetry, and power spectrum are extracted from five bands of each EEG channel. We propose a new feature selection algorithm, which selects features from appropriate EEG frequency band based on classification accuracy. Three classifiers i.e., support vector machine, the Naive Bayes, and multi-layer perceptron are used to classify stress level of the participants. It is evident from our results that EEG recording during the pre-activity phase is better for classifying the perceived stress. An accuracy of [Formula: see text] and [Formula: see text] is achieved for two- and three-class stress classification, respectively, while utilizing five groups of features from theta band. Our proposed feature selection algorithm is compared with existing algorithms and gives better classification results.

Here, a hybrid feature descriptor‐based method is proposed to recognise human emotions from their facial expressions. A combination of spatial bag of features (SBoFs) with spatial scale‐invariant feature transform (SBoF‐SSIFT), and SBoFs with spatial speeded up robust transform are utilised to improve the ability to recognise facial expressions. For classification of emotions, K ‐nearest neighbour and support vector machines (SVMs) with linear, polynomial, and radial basis function kernels are applied. SBoFs descriptor generates a fixed length feature vector for all sample images irrespective of their size. Spatial SIFT and SURF features are independent of scaling, rotation, translation, projective transforms, and partly to illumination changes. A modified form of bag of features (BoFs) is employed by involving feature's spatial information for facial emotion recognition. The proposed method differs from conventional methods that are used for simple object categorisation without using spatial information. Experiments have been performed on extended Cohn–Kanade (CK+) and Japanese female facial expression (JAFFE) data sets. SBoF‐SSIFT with SVM resulted in a recognition accuracy of 98.5% on CK+ and 98.3% on JAFFE data set. Images are resized through selective pre‐processing, thereby retaining only the information of interest and reducing computation time.

Optic disc is one of the fundamental regions located in the internal retina that helps ophthalmologists in analysis and early diagnosis of many retinal diseases such as optic atrophy, optic neuritis, papilledema, ischemic optic neuropathy, glaucoma and diabetic retinopathy. An accurate and early diagnosis requires an accurate optic disc examination. Presence of different retinal abnormalities and non-uniform illumination make optic disc localization a challenging task. There is a need to detect and localize optic disc from fundus images with high accuracy to make the diagnosis using Computer Aided Systems developed for ophthalmic disease diagnosis more reliable. Proposed algorithm provides a novel optic disc localization and segmentation technique that detects multiple candidate optic disc regions from fundus image using enhancement and segmentation. The proposed system then extracts a hybrid feature set for each candidate region consisting of vessel based and intensity based features which are finally fed to SVM classifier. Final decision of Optic disc region is done after computing Manhattan distance from the mean of training data feature matrix. The evaluation of proposed system has been done on publicly available datasets and one local dataset and results shows the validity of proposed system.