Clinicodemographic profile and outcome of tuberculosis treatment in TB-HIV co-infected patients receiving daily ATT under a single window TB/HIV services delivery initiative.

The risk of death in HIV-TB coinfected individuals is far greater than in HIV-only patients. It is critical to provide timely and appropriate therapy in HIV-TB coinfected patients in order to reduce morbidity and mortality. The purpose of this study was to evaluate the clinical presentation and outcome of TB treatment in HIV-TB co-infected patients receiving daily anti-tubercular therapy (ATT) and concurrent antiretroviral therapy (ART) at a tertiary respiratory care centre in New Delhi, India. The research was cross-sectional, observational, and hospital-based A. From September 2018 to August 2019, a total of 53 patients with HIV-TB coinfection were enrolled at the Institute's ART centre. Patients were evaluated with a structured proforma. Data were evaluated using SPSS version 23.0 and p-value of less than 0.05 was considered statistically significant. Among the patients enrolled, the mean age was 35.98 years. Among the patients enrolled, 56.6% patients had EPTB, 32% had PTB and 11.3% had both PTB and EPTB. The majority of the enrolled patients (n=46, 86.7%) had favourable TB treatment outcomes, while 13.3% (n=7) had unfavourable outcome [including death (n=5) and loss to follow up (n=2)]. During the study and follow-up period, no patients transferred out or relapsed. In univariate analysis, low SES, bedridden functional status, low BMI, anaemia, hypoalbuminemia, and a low CD-4 cell count (<100 cells/mm3 were significantly associated with an unfavourable outcome. Bedridden functional status (p=0.002), anaemia (p=0.040), and low BMI (p<0.001) were independently associated with a poor outcome. Adequate disease knowledge and health education can be very beneficial in reducing morbidity and mortality. Early ART in combination with ATT can reduce mortality in TB-HIV co-infected patients.

Deep neural network and random forest hybrid architecture for learning to detect retinal vessels in fundus images.

Vision impairment due to pathological damage of the retina can largely be prevented through periodic screening using fundus color imaging. However the challenge with large-scale screening is the inability to exhaustively detect fine blood vessels crucial to disease diagnosis. In this work we present a computational imaging framework using deep and ensemble learning based hybrid architecture for reliable detection of blood vessels in fundus color images. A deep neural network (DNN) is used for unsupervised learning of vesselness dictionaries using sparse trained denoising auto-encoders (DAE), followed by supervised learning of the DNN response using a random forest for detecting vessels in color fundus images. In experimental evaluation with the DRIVE database, we achieve the objective of vessel detection with max. avg. accuracy of 0.9327 and area under ROC curve of 0.9195.