Artificial Intelligence and Infectious Disease Imaging.

The mass production of the graphics processing unit and the coronavirus disease 2019 (COVID-19) pandemic have provided the means and the motivation, respectively, for rapid developments in artificial intelligence (AI) and medical imaging techniques. This has led to new opportunities to improve patient care but also new challenges that must be overcome before these techniques are put into practice. In particular, early AI models reported high performances but failed to perform as well on new data. However, these mistakes motivated further innovation focused on developing models that were not only accurate but also stable and generalizable to new data. The recent developments in AI in response to the COVID-19 pandemic will reap future dividends by facilitating, expediting, and informing other medical AI applications and educating the broad academic audience on the topic. Furthermore, AI research on imaging animal models of infectious diseases offers a unique problem space that can fill in evidence gaps that exist in clinical infectious disease research. Here, we aim to provide a focused assessment of the AI techniques leveraged in the infectious disease imaging research space, highlight the unique challenges, and discuss burgeoning solutions.

Circulating Cell-free DNA in Serum as a Marker for the Early Detection of Tumor Recurrence in Breast Cancer Patients.

BACKGROUND/AIM: Circulating cell-free DNA (cfDNA) isolated from serum by noninvasive procedures can serve as a potential biomarker for the early detection of many cancers. The aim of this study was to implement a simple, yet effective quantitative method for measuring the cfDNA in serum and to investigate the relationship between cfDNA and the occurrence of recurrence in breast cancer (BrCa) patients. PATIENTS AND METHODS: A total of 240 cases were selected, which comprised different subtypes of BrCa patients and control individuals. We selected 20 serum samples from patients which showed recurrence after 4-7 years of disease-free survival. SYBR green was used as a reporter molecule to estimate the amount of cfDNA in these serum samples. RESULTS: A global Wilcoxon analysis was performed to compare the cfDNA abundance between non-recurrent and recurrent patients. The amount of cfDNA was higher in recurrent patients (recurrent vs. non-recurrent ratio=1.3; p=0.03; AUC=0.76) compared to non-recurrent patients. The data between normal/healthy controls and non-recurrent patients indicated no significant differences (n=20 in each group, healthy to non-recurrent ratio=1.03; p=0.20; AUC=0.61). CONCLUSION: We implemented a straightforward one-step technique to measure the amount of cfDNA in serum, which can translate into a clinical diagnostic tool in the near future. The high levels of cfDNA in the serum of recurrent BrCa patients compared to non-recurrent BrCa patients indicates a possible uncovered role for circulating genetic information, which either contributes to the cancer recurrence phenomenon or at the very least, serves as an identifier for the potential of recurrence.

Digital Omicron detection using unscripted voice samples from social media.

The success of artificial intelligence in clinical environments relies upon the diversity and availability of training data. In some cases, social media data may be used to counterbalance the limited amount of accessible, well-curated clinical data, but this possibility remains largely unexplored. In this study, we mined YouTube to collect voice data from individuals with self-declared positive COVID-19 tests during time periods in which Omicron was the predominant variant1,2,3, while also sampling non-Omicron COVID-19 variants, other upper respiratory infections (URI), and healthy subjects. The resulting dataset was used to train a DenseNet model to detect the Omicron variant from voice changes. Our model achieved 0.85/0.80 specificity/sensitivity in separating Omicron samples from healthy samples and 0.76/0.70 specificity/sensitivity in separating Omicron samples from symptomatic non-COVID samples. In comparison with past studies, which used scripted voice samples, we showed that leveraging the intra-sample variance inherent to unscripted speech enhanced generalization. Our work introduced novel design paradigms for audio-based diagnostic tools and established the potential of social media data to train digital diagnostic models suitable for real-world deployment.