In this work, Uncertainty-Guided Gradual Learning (UGGL) is proposed for one-shot segmentation of slice-stacks, where only a single labeled slice is available. UGGL extends regular Gradual Learning (GL) with ensemble-based consensus for adaptive sample selection and entropy-based uncertainty weighting during student training. A k-head teacher ensemble is trained on the single ground-truth slice to generate pseudo labels (PLs), uncertainty, and consensus maps. UGGL then selects neighboring slices with high mean consensus (excluding boundary regions) and iteratively trains a student ensemble, prioritizing reliable PLs to prevent error propagation. The evaluation was conducted on the Calgary-Campinas MRI dataset for brain segmentation and compared UGGL against GL with fixed step rates and guided teacher–student (TS) training. Trained from one labeled slice per volume, UGGL improved mean Intersection over Union (IoU) by approximately 0.10 over the one-slice baseline, reducing the remaining gap to fully supervised training to approximately 0.02 IoU, and achieved higher precision and recall with lower variance than both GL and TS. Cross-scanner experiments further indicated superior generalization relative to GL and TS, with the strongest gains on challenging slices. Qualitative 2D and 3D results showed cleaner masks and reduced boundary errors. These properties make UGGL a broadly applicable method for one-shot segmentation in real-world settings.
This study investigates the concentrations of potentially toxic elements (PTEs) in sediments from the Sundarbans mangrove region, India. It represents the unique first coast-wide comprehensive assessment of intertidal sediment-associated PTEs across the Indian Sundarbans, integrating GIS-based spatial analysis, Positive Matrix Factorization (PMF) based source apportionment, multiple pollution indices, and human health risk assessment. A total of 50 samples (n = 3 replicates per site) were collected from different intertidal locations along the coast and analyzed for Al, Ba, Cd, Cr, Co, Cu, Fe, Pb, Mn, V, and Zn using Inductively Coupled Plasma- Optical Emission Spectroscopy (ICP-OES). Sediment contamination was evaluated using multiple pollution indices such as contamination factor (CF), enrichment factor (EF), geoaccumulation index (Igeo), ecological risk index (ERI), pollution load index (PLI), and improved Nemerow index (INI). Unlike conventional multivariate approaches, PMF quantitatively resolved the major pollution sources and estimated their relative contributions, providing a more robust basis for source identification and pollution management. The mean concentrations of PTEs followed the descending order: Fe > Al > Mn > Cr > Ba > V > Zn > Cu > Co > Pb > Cd. Among these, Fe (65,545.72 mg/kg), Al (3975.73 mg/kg), Mn (2980.39 mg/kg), and Cr (658.30 mg/kg) were the most abundant elements. The Igeo values (−6.21 to 5.91) and EF values (0.01 to 37.50) suggest significant enrichment of Cd, Cr, and Pb in the sediments. The overall PLI was 1.26, with Cd and Pb identified as the primary contributors to sediment contamination. Health risk assessment indicated that the hazard index (HI) for Cr (2.72) in children exceeds the acceptable safety threshold, suggesting potential non-carcinogenic health risks. The total carcinogenic risk (TCR) for children ranged from 1.15 × 10−5 to 3.67 × 10–3, and for adults, 6.88 × 10−6 to 5.50 × 10−4. Overall, this study, provides a scientific framework for identifying contamination hotspots, pollution sources, and supporting evidence-based environmental management and policy interventions.
