Differentiation of urothelial carcinoma in histopathology images using deep learning and visualization.

Artificial Intelligence is a tool poised to transform healthcare, with use in diagnostics and therapeutics. The widespread use of digital pathology has been due to the advent of whole slide imaging. Cheaper storage for digital images, along with unprecedented progress in artificial intelligence, have paved the synergy of these two fields. This has pushed the limits of traditional diagnosis using light microscopy, from a more subjective to a more objective method of looking at cases, incorporating grading too. The grading of histopathological images of urothelial carcinoma of the urinary bladder is important with direct implications for surgical management and prognosis. In this study, the aim is to classify urothelial carcinoma into low and high grade based on the WHO 2016 classification. The hematoxylin and eosin-stained transurethral resection of bladder tumor (TURBT) samples of both low and high grade non-invasive papillary urothelial carcinoma were digitally scanned. Patches were extracted from these whole slide images to feed into a deep learning (Convolution Neural Network: CNN) model. Patches were segregated if they had tumor tissue and only included for model training if a threshold of 90% of tumor tissue per patch was seen. Various parameters of the deep learning model, known as hyperparameters, were optimized to get the best accuracy for grading or classification into low- and high-grade urothelial carcinoma. The model was robust with an overall accuracy of 90% after hyperparameter tuning. Visualization in the form of a class activation map using Grad-CAM was done. This indicates that such a model can be used as a companion diagnostic tool for grading of urothelial carcinoma. The probable causes of this accuracy are summarized along with the limitations of this study and future work possible.

Cardiolipin-mediated PPARγ S112 phosphorylation impairs IL-10 production and inflammation resolution during bacterial pneumonia.

Bacterial pneumonia is a global healthcare burden, and unwarranted inflammation is suggested as an important cause of mortality. Optimum levels of the anti-inflammatory cytokine IL-10 are essential to reduce inflammation and improve survival in pneumonia. Elevated levels of the mitochondrial-DAMP cardiolipin (CL), reported in tracheal aspirates of pneumonia patients, have been shown to block IL-10 production from lung MDSCs. Although CL-mediated K107 SUMOylation of PPARγ has been suggested to impair this IL-10 production, the mechanism remains elusive. We identify PIAS2 to be the specific E3-SUMOligase responsible for this SUMOylation. Moreover, we identify a concomitant CL-mediated PPARγ S112 phosphorylation, mediated by JNK-MAPK, to be essential for PIAS2 recruitment. Furthermore, using a clinically tested peptide inhibitor targeting JNK-MAPK, we blocked these post-translational modifications (PTMs) of PPARγ and rescued IL-10 expression, improving survival in murine pneumonia models. Thus, we explore the mechanism of mito-DAMP-mediated impaired lung inflammation resolution and propose a therapeutic strategy targeting PPARγ PTMs.

A study of organisms causing surgical site infections and their antimicrobial susceptibility in a tertiary care government hospital.

BACKGROUND: Surgical site infection (SSI) is one of the most common postoperative complication and causes significant postoperative morbidity and mortality. PATIENTS: A prospective study was carried out in a total of 100 patients operated for clean and clean-contaminated surgeries from department of orthopedics, surgery and obstetrics & gynecology. MATERIALS AND METHODS: Relevant details were noted in clinical history. Each patient was followed from the time of admission till discharge from the hospital and also for 30 days postoperatively (CDC, 1999). The identification of the infecting organism was done by staining, and culture and antibiotic susceptibility by Disc Diffusion method. RESULTS: Out of 100 patients, 32 patients got infected post-operatively. Staphylococcus aureus was the most common organism isolated. None of the strains were Methicillin resistant. Drug resistance was widespread, especially in Enterobacteriaceae, where the Cefotaxime resistant strains of Escherichia coli and Klebsiella pneumoniae were ESBL producing. Another concern in recent times is the isolation of Acinetobacter from surgical wounds. Various patient factors and hospital protocol were analyzed with regard to the treatment outcome. Judicious use of antibiotics along with evidence-based medicine is the need of the hour to stop the rise of these superbugs.