Learning to Generalize Towards Unseen Domains via a Content-Aware Style Invariant Model for Disease Detection From Chest X-Rays.

Performance degradation due to distribution discrepancy is a longstanding challenge in intelligent imaging, particularly for chest X-rays (CXRs). Recent studies have demonstrated that CNNs are biased toward styles (e.g., uninformative textures) rather than content (e.g., shape), in stark contrast to the human vision system. Radiologists tend to learn visual cues from CXRs and thus perform well across multiple domains. Motivated by this, we employ the novel on-the-fly style randomization modules at both image (SRM-IL) and feature (SRM-FL) levels to create rich style perturbed features while keeping the content intact for robust cross-domain performance. Previous methods simulate unseen domains by constructing new styles via interpolation or swapping styles from existing data, limiting them to available source domains during training. However, SRM-IL samples the style statistics from the possible value range of a CXR image instead of the training data to achieve more diversified augmentations. Moreover, we utilize pixel-wise learnable parameters in the SRM-FL compared to pre-defined channel-wise mean and standard deviations as style embeddings for capturing more representative style features. Additionally, we leverage consistency regularizations on global semantic features and predictive distributions from with and without style-perturbed versions of the same CXR to tweak the model's sensitivity toward content markers for accurate predictions. Our proposed method, trained on CheXpert and MIMIC-CXR datasets, achieves 77.32±0.35, 88.38±0.19, 82.63±0.13 AUCs(%) on the unseen domain test datasets, i.e., BRAX, VinDr-CXR, and NIH chest X-ray14, respectively, compared to 75.56±0.80, 87.57±0.46, 82.07±0.19 from state-of-the-art models on five-fold cross-validation with statistically significant results in thoracic disease classification.

Trace elements in rice grain and agricultural soils: assessment of health risk of inhabitants near a former secondary lead smelter in Khulna, Bangladesh.

Ingestion of food grain grown in metal-contaminated soils may cause serious effects on human health. This study assessed the concentrations of Pb, As, Cd and Zn in agricultural soils and in rice grains near a former secondary lead smelter in Khulna, Bangladesh. It analyzed 29 samples of surface soil and rice grain collected around 500 m of the smelter. Contamination factor (Cf), pollution load index and total hazard quotient (THQ) were calculated to determine ecological and human health risks. Cd was not detected in any of the samples. For the soil samples, medians of the concentrations of Pb, As and Zn were 109, 6.2 and 514 mg/kg, respectively. For the rice grain samples, medians of the concentrations of Pb, As and Zn were 4, 1.4 and 25 mg/kg fw, respectively. Medians of the concentrations of Pb and As in rice grain were higher compared to their maximum allowable limit (0.2 mg/kg), which indicate potential health risks to inhabitants near the Pb smelter. The mean values of Cf for Pb, As, and Zn were, respectively, 11.6, 2.1 and 7.4. For Pb, around 41% of the samples had Cf > 6 indicating very strong contamination. THQ values for Pb and As were greater than 1.0, which evinces the health hazards of these trace elements. Measures should be taken to prevent trace elements exposure from Pb smelter in the study area.