RESUMO
Cancer outcomes are poor in resource-limited countries owing to high costs and insufficient pathologist-population ratio. The advent of digital pathology has assisted in improving cancer outcomes, however, Whole Slide Image scanners are expensive and not affordable in low-income countries. Microscope-acquired images on the other hand are cheap to collect and can be more viable for automation of cancer detection. In this study, we propose LCH-Network, a novel method to identify the cancer mitotic count from microscope-acquired images. We introduced Label Mix, and also synthesized images using GANs to handle data imbalance. Moreover, we applied progressive resolution to handle different image scales for mitotic localization. We achieved F1-Score of 0.71 and outperformed other existing techniques. Our findings enable mitotic count estimation from microscopic images with a low-cost setup. Clinically, our method could help avoid presumptive treatment without a confirmed cancer diagnosis.
RESUMO
BACKGROUND: Cutaneous leishmaniasis (CL) remains a prioritized neglected tropical disease. CL novel presentations call for updating its features. METHODS: A multiregional cohort of 396 patients with confirmed CL was reviewed. Lesion's clinical stage and eruption type were assigned. Disease was considered as extensive if numerous (≥5), large (>3 cm), disfiguring, threatening vital sensory organs, and/or older than 12 months. Microscopically, Ackerman's inflammatory pattern, Ridley's pattern (RP), and parasitic index (PI) were recorded. Microscopic variables pertaining to the organisms, epidermis, and host's inflammatory response were also assessed. All cases were confirmed and speciated molecularly. RESULTS: In our region, 71.8% of cases showed extensive disease with 15.7% exceeding 12 months duration. Leishmania tropica accounted for 91.3% of cases while Leishmania major constituted 8.7% and presented solely as dry lesions. The dominant inflammatory composite consisted of plasma cells, lymphocytes, and histiocytes. Granulomatous inflammation was present in 55.5%. Most cases showed interface changes (72.7%), spongiosis (75.3%), and marked epidermal hyperplasia (63.9%). Transepidermal elimination of organisms was present in 29.2% of cases. None of traditional classification patterns (clinical stage, microscopic pattern, and RP) showed the predicted linear correlation with lesion age. High and low PI levels correlated with early and healing microscopic patterns, respectively, but did not correlate with the corresponding RPs. PI was bimodal with peaks at 3-6 and 9-12 months. CONCLUSION: Cutaneous leishmaniasis is an evolving disease defying the traditional prediction classifications. Our study sets the ground for adopting updated clinical courses, microscopic presentation, and species mapping.