Search

Background: The exact location of skin lesions is key in clinical dermatology. On one hand, it supports differential diagnosis (DD) since most skin conditions have specific predilection sites. On the other hand, location matters for dermatosurgical interventions. In practice, lesion evaluation is not well standardized and anatomical descriptions vary or lack altogether. Automated determination of anatomical location could benefit both situations. Objective: Establish an automated method to determine anatomical regions in clinical patient pictures and evaluate the gain in DD performance of a deep learning model (DLM) when trained with lesion locations and images. Methods: Retrospective study based on three datasets: macro‐anatomy for the main body regions with 6000 patient pictures partially labelled by a student, micro‐anatomy for the ear region with 182 pictures labelled by a student and DD with 3347 pictures of 16 diseases determined by dermatologists in clinical settings. For each dataset, a DLM was trained and evaluated on an independent test set. The primary outcome measures were the precision and sensitivity with 95% CI. For DD, we compared the performance of a DLM trained with lesion pictures only with a DLM trained with both pictures and locations. Results: The average precision and sensitivity were 85% (CI 84–86), 84% (CI 83–85) for macro‐anatomy, 81% (CI 80–83), 80% (CI 77–83) for micro‐anatomy and 82% (CI 78–85), 81% (CI 77–84) for DD. We observed an improvement in DD performance of 6% (McNemar test P‐value 0.0009) for both average precision and sensitivity when training with both lesion pictures and locations. Conclusion: Including location can be beneficial for DD DLM performance. The proposed method can generate body region maps from patient pictures and even reach surgery relevant anatomical precision, e.g. the ear region. Our method enables automated search of large clinical databases and make targeted anatomical image retrieval possible. [ABSTRACT FROM AUTHOR]

Objectives To determine frequency and syndrome specificity of novel and known nervous system (NS)-directed antibodies in a large, unbiased cohort of SLE patients in the Swiss SLE Cohort Study. Methods This retrospective pilot study included 174 patients in a cross-sectional and 102 in a longitudinal study. Antibodies against 12 NS antigens [myelin oligodendrocyte glycoprotein (MOG), neurofascin 186 (NF186), aquaporin-4 (AQP4), N -methyl-D-aspartate receptor (subunit NR1) (NMDAR-NR1), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (subunits 1 and 2) (AMPAR1/2), gamma-aminobutyric acid B receptor (subunits B1 and B2) (GABABR1/2), glutamate decarboxylase 65 (GAD65), glycine receptor (GlyR), contactin-associated protein-like 2 (CASPR2), leucine-rich glioma-inactivated 1 (LGI1), metabotropic glutamate receptor 5 (mGluR5) and dipeptidyl-peptidase-like protein 6 (DPPX)] were screened with validated cell-based assays and correlated with clinical and diagnostic findings. Results Twenty-three of one hundred and seventy-four (13.2%) patients harboured antibodies against MOG (n = 14), NF186 (n = 6), GAD65 (n = 2), AQP4 and GlyR (n = 1). Anti-MOG antibodies were most frequently found in the cohort (8%). Thirteen of the anti-NS antibody-positive patients showed clinical symptoms of NS involvement, a subgroup of which (n = 8) resembled the syndrome associated with the antibody. Nine patients harboured antibodies without neurological symptoms and one patient was lost to follow-up. The frequency of NPSLE was significantly higher in the anti-NS antibody-positive patients (13/23, 56.5%: MOG 6/14, 42.9%; NF186 5/6, 83.3%; GAD65 2/2, 100%; AQP4/GlyR 0/1, 0%) compared with the antibody-negative cohort (21/151, 13.9%) (chi-square test, P < 0.0001). Conclusion Anti-NS antibodies, most prevalently anti-MOG antibodies, are significantly associated with NPSLE and manifest with the distinct neurological syndrome associated with the antibody in a subgroup. Follow-up studies in large, independent cohorts will reveal whether these anti-NS antibodies could serve as a diagnostic and prognostic biomarker for NPSLE and enable tailored treatment decisions in this challenging and diverse patient cohort. [ABSTRACT FROM AUTHOR]