The application of deep learning to pathology assumes the existence of digital whole slide images of pathology slides. However, slide digitization is bottlenecked by the high cost of precise motor ...stages in slide scanners that are needed for position information used for slide stitching. We propose GloFlow, a two-stage method for creating a whole slide image using optical flow-based image registration with global alignment using a computationally tractable graph-pruning approach. In the first stage, we train an optical flow predictor to predict pairwise translations between successive video frames to approximate a stitch. In the second stage, this approximate stitch is used to create a neighborhood graph to produce a corrected stitch. On a simulated dataset of video scans of WSIs, we find that our method outperforms known approaches to slide-stitching, and stitches WSIs resembling those produced by slide scanners.
: The objectives were to assess whether any deaths reported among perinatally exposed, uninfected, or indeterminate children were consistent with mitochondrial dysfunction. and to characterize ...perinatal exposure to antiretrovirals among children born in the last five years and reported to perinatal HIV surveillance. Population‐based HIV/AIDS surveillance data on perinatally exposed children born in 1993 through 1998 from 32 states with HIV reporting and from a special HIV surveillance project in Los Angeles County and in 22 hospitals in New York City were used. The classifications of exposure and deaths were consistent with the investigation of deaths across all US cohorts. Deaths were ascertained from recent matches with death registries in each state. Causes of death were ascertained from death certificates, autopsy records when available, and medical records. None of the 98 deaths (1.1%) among 9067 perinatally exposed uninfected or indeterminate children born from 1993 through 1998 and reported through pediatric HIV surveillance died of conditions that were consistent with mitochondrial dysfunction. This included 679 children exposed to zidovudine (ZDV) and 3TC, 277 exposed to other antiretroviral combinations, 4512 exposed to ZDV alone, 927 with no antiretroviral exposure, and 2672 with unknown exposure‐1128 of whom were born before March 1994 and were unlikely to have been exposed to ZDV. No deaths attributable to mitochondrial dysfunction were found through this evaluation of population‐based HIV surveillance data. Long‐term follow‐up of antiretroviral‐exposed children has been recommended by the Public Health Service. This evaluation highlights the contribution of population‐based surveillance to the evaluation of potential toxicities associated with maternal antiretroviral use.