Alanine transaminase is predominantly increased in the active phase of anti-HMGCR myopathy

Autoantibodies against 3?hydroxy-3-methylglutaryl-CoA reductase (HMGCR) and the signal recognition particle (SRP) are representative antibodies causing immune-mediated necrotizing myopathies (IMNM), called as anti-HMGCR and anti-SRP myopathies, respectively. Here, we analyzed the differences in routine blood test results between 56 anti-HMGCR and 77 anti-SRP myopathy patients. A higher alanine transaminase (ALT) level and a lower aspartate transaminase (AST)/ALT ratio were observed in anti-HMGCR myopathy patients [ALT, 265.7 ± 213.3 U/L (mean ± standard deviation); AST/ALT ratio, 0.88 ± 0.32] than in anti-SRP-myopathy patients (ALT, 179.3 ± 111.2 U/L, p < 0.05; AST/ALT ratio, 1.28 ± 0.40, p < 0.01). In the active phase, anti-HMGCR myopathy often showed ALT predominance, whereas anti-SRP myopathy often showed AST predominance. In addition, there were differences in erythrocyte sedimentation rate (ESR), total cholesterol (TChol) level, and high-density lipoprotein (HDL) level between anti-HMGCR and anti-SRP myopathies (ESR: HMGCR, 24.4 ± 20.8 mm/1 h; SRP, 35.7 ± 26.7 mm/1 h, p = 0.0334; TChol: HMGCR, 226.7 ± 36.6 mg/dL; SRP, 207.6 ± 40.8 mg/dL, p = 0.0163; HDL: HMGCR, 58.4 ± 13.9 mg/dL; SRP, 46.2 ± 17.3 mg/dL, p < 0.01). Additional studies on the differences in routine blood test results may further reveal the pathomechanisms of IMNM.     

Novel pathogenic variants in an Indian cohort with epidermolysis bullosa: Expanding the genotypic spectrum

BackgroundEpidermolysis bullosa (EB) is a genodermatosis characterized by skin fragility and blisters with variable severity. Patients with Dystrophic EB (DEB) or Junctional EB (JEB) mainly present to clinic due to greater functional impairment. Pathogenic sequence variations in COL7A1 are implicated in DEB.ObjectiveWe have tried to decipher the molecular spectrum and genotype phenotype correlation of 21 Indian patients with EB.MethodsNext generation sequencing (NGS) was performed to determine the pathogenic variants. Sanger sequencing was also done for validation of the variants in eleven individuals.ResultsPathogenic variants were detected in 20 individuals (diagnostic yield of 95%). Majority of them (90%) had sequence variation in COL7A1 while two had pathogenic variants in ITGB4 and KRT14 respectively. Out of the 18 patients confirmed to have DEB, 3 had Dominant DEB (DDEB) whereas 15 patients had Recessive DEB (RDEB). Amongst 23 sequence variations identified, 12 were found to be novel (3 were missense, 5 were premature termination codon variants while 4 were splice-site changes).ConclusionGenotype phenotype correlation was noted with milder manifestations in those with dominant inheritance types. Exact molecular diagnosis can be ascertained by NGS in majority of cases.     

Quantifying and leveraging predictive uncertainty for medical image assessment

The interpretation of medical images is a challenging task, often complicated by the presence of artifacts, occlusions, limited contrast and more. Most notable is the case of chest radiography, where there is a high inter-rater variability in the detection and classification of abnormalities. This is largely due to inconclusive evidence in the data or subjective definitions of disease appearance. An additional example is the classification of anatomical views based on 2D Ultrasound images. Often, the anatomical context captured in a frame is not sufficient to recognize the underlying anatomy. Current machine learning solutions for these problems are typically limited to providing probabilistic predictions, relying on the capacity of underlying models to adapt to limited information and the high degree of label noise. In practice, however, this leads to overconfident systems with poor generalization on unseen data. To account for this, we propose a system that learns not only the probabilistic estimate for classification, but also an explicit uncertainty measure which captures the confidence of the system in the predicted output. We argue that this approach is essential to account for the inherent ambiguity characteristic of medical images from different radiologic exams including computed radiography, ultrasonography and magnetic resonance imaging. In our experiments we demonstrate that sample rejection based on the predicted uncertainty can significantly improve the ROC-AUC for various tasks, e.g., by 8% to 0.91 with an expected rejection rate of under 25% for the classification of different abnormalities in chest radiographs. In addition, we show that using uncertainty-driven bootstrapping to filter the training data, one can achieve a significant increase in robustness and accuracy. Finally, we present a multi-reader study showing that the predictive uncertainty is indicative of reader errors.