Advancements in the future of automating micromanipulation techniques in the IVF laboratory using deep convolutional neural networks.

PURPOSE: To determine if deep learning artificial intelligence algorithms can be used to accurately identify key morphologic landmarks on oocytes and cleavage stage embryo images for micromanipulation procedures such as intracytoplasmic sperm injection (ICSI) or assisted hatching (AH). METHODS: Two convolutional neural network (CNN) models were trained, validated, and tested over three replicates to identify key morphologic landmarks used to guide embryologists when performing micromanipulation procedures. The first model (CNN-ICSI) was trained (n = 13,992), validated (n = 1920), and tested (n = 3900) to identify the optimal location for ICSI through polar body identification. The second model (CNN-AH) was trained (n = 13,908), validated (n = 1908), and tested (n = 3888) to identify the optimal location for AH on the zona pellucida that maximizes distance from healthy blastomeres. RESULTS: The CNN-ICSI model accurately identified the polar body and corresponding optimal ICSI location with 98.9% accuracy (95% CI 98.5-99.2%) with a receiver operator characteristic (ROC) with micro and macro area under the curves (AUC) of 1. The CNN-AH model accurately identified the optimal AH location with 99.41% accuracy (95% CI 99.11-99.62%) with a ROC with micro and macro AUCs of 1. CONCLUSION: Deep CNN models demonstrate powerful potential in accurately identifying key landmarks on oocytes and cleavage stage embryos for micromanipulation. These findings are novel, essential stepping stones in the automation of micromanipulation procedures.

Estimated US Infection- and Vaccine-Induced SARS-CoV-2 Seroprevalence Based on Blood Donations, July 2020-May 2021.

Importance: People who have been infected with or vaccinated against SARS-CoV-2 have reduced risk of subsequent infection, but the proportion of people in the US with SARS-CoV-2 antibodies from infection or vaccination is uncertain. Objective: To estimate trends in SARS-CoV-2 seroprevalence related to infection and vaccination in the US population. Design, Setting, and Participants: In a repeated cross-sectional study conducted each month during July 2020 through May 2021, 17 blood collection organizations with blood donations from all 50 US states; Washington, DC; and Puerto Rico were organized into 66 study-specific regions, representing a catchment of 74% of the US population. For each study region, specimens from a median of approximately 2000 blood donors were selected and tested each month; a total of 1 594 363 specimens were initially selected and tested. The final date of blood donation collection was May 31, 2021. Exposure: Calendar time. Main Outcomes and Measures: Proportion of persons with detectable SARS-CoV-2 spike and nucleocapsid antibodies. Seroprevalence was weighted for demographic differences between the blood donor sample and general population. Infection-induced seroprevalence was defined as the prevalence of the population with both spike and nucleocapsid antibodies. Combined infection- and vaccination-induced seroprevalence was defined as the prevalence of the population with spike antibodies. The seroprevalence estimates were compared with cumulative COVID-19 case report incidence rates. Results: Among 1 443 519 specimens included, 733 052 (50.8%) were from women, 174 842 (12.1%) were from persons aged 16 to 29 years, 292 258 (20.2%) were from persons aged 65 years and older, 36 654 (2.5%) were from non-Hispanic Black persons, and 88 773 (6.1%) were from Hispanic persons. The overall infection-induced SARS-CoV-2 seroprevalence estimate increased from 3.5% (95% CI, 3.2%-3.8%) in July 2020 to 20.2% (95% CI, 19.9%-20.6%) in May 2021; the combined infection- and vaccination-induced seroprevalence estimate in May 2021 was 83.3% (95% CI, 82.9%-83.7%). By May 2021, 2.1 SARS-CoV-2 infections (95% CI, 2.0-2.1) per reported COVID-19 case were estimated to have occurred. Conclusions and Relevance: Based on a sample of blood donations in the US from July 2020 through May 2021, vaccine- and infection-induced SARS-CoV-2 seroprevalence increased over time and varied by age, race and ethnicity, and geographic region. Despite weighting to adjust for demographic differences, these findings from a national sample of blood donors may not be representative of the entire US population.

Adaptive adversarial neural networks for the analysis of lossy and domain-shifted datasets of medical images.

In machine learning for image-based medical diagnostics, supervised convolutional neural networks are typically trained with large and expertly annotated datasets obtained using high-resolution imaging systems. Moreover, the network's performance can degrade substantially when applied to a dataset with a different distribution. Here, we show that adversarial learning can be used to develop high-performing networks trained on unannotated medical images of varying image quality. Specifically, we used low-quality images acquired using inexpensive portable optical systems to train networks for the evaluation of human embryos, the quantification of human sperm morphology and the diagnosis of malarial infections in the blood, and show that the networks performed well across different data distributions. We also show that adversarial learning can be used with unlabelled data from unseen domain-shifted datasets to adapt pretrained supervised networks to new distributions, even when data from the original distribution are not available. Adaptive adversarial networks may expand the use of validated neural-network models for the evaluation of data collected from multiple imaging systems of varying quality without compromising the knowledge stored in the network.

Mobile Health (mHealth) Viral Diagnostics Enabled with Adaptive Adversarial Learning.

Deep-learning (DL)-based image processing has potential to revolutionize the use of smartphones in mobile health (mHealth) diagnostics of infectious diseases. However, the high variability in cellphone image data acquisition and the common need for large amounts of specialist-annotated images for traditional DL model training may preclude generalizability of smartphone-based diagnostics. Here, we employed adversarial neural networks with conditioning to develop an easily reconfigurable virus diagnostic platform that leverages a dataset of smartphone-taken microfluidic chip photos to rapidly generate image classifiers for different target pathogens on-demand. Adversarial learning was also used to augment this real image dataset by generating 16,000 realistic synthetic microchip images, through style generative adversarial networks (StyleGAN). We used this platform, termed smartphone-based pathogen detection resource multiplier using adversarial networks (SPyDERMAN), to accurately detect different intact viruses in clinical samples and to detect viral nucleic acids through integration with CRISPR diagnostics. We evaluated the performance of the system in detecting five different virus targets using 179 patient samples. The generalizability of the system was confirmed by rapid reconfiguration to detect SARS-CoV-2 antigens in nasal swab samples (n = 62) with 100% accuracy. Overall, the SPyDERMAN system may contribute to epidemic preparedness strategies by providing a platform for smartphone-based diagnostics that can be adapted to a given emerging viral agent within days of work.