Predictive mutation signature of immunotherapy benefits in NSCLC based on machine learning algorithms.

Background: Developing prediction tools for immunotherapy approaches is a clinically important and rapidly emerging field. The routinely used prediction biomarker is inaccurate and may not adequately utilize large amounts of medical data. Machine learning is a promising way to predict the benefit of immunotherapy from individual data by individuating the most important features from genomic data and clinical characteristics. Methods: Machine learning was applied to identify a list of candidate genes that may predict immunotherapy benefits using data from the published cohort of 853 patients with NSCLC. We used XGBoost to capture nonlinear relations among many mutation genes and ICI benefits. The value of the derived machine learning-based mutation signature (ML-signature) on immunotherapy efficacy was evaluated and compared with the tumor mutational burden (TMB) and other clinical characteristics. The predictive power of ML-signature was also evaluated in independent cohorts of patients with NSCLC treated with ICI. Results: We constructed the ML-signature based on 429 (training/validation = 8/2) patients who received immunotherapy and extracted 88 eligible predictive genes. Additionally, we conducted internal and external validation with the utility of the OAK+POPLAR dataset and independent cohorts, respectively. This ML-signature showed the enrichment in immune-related signaling pathways and compared to TMB, ML-signature was equipped with favorable predictive value and stratification. Conclusion: Previous studies proposed no predictive difference between original TMB and modified TMB, and original TMB contains some genes with no predictive value. To demonstrate that fewer genetic tests are sufficient to predict immunotherapy efficacy, we used machine learning to screen out gene panels, which are used to calculate TMB. Therefore, we obtained the 88-gene panel, which showed the favorable prediction performance and stratification effect compared to the original TMB.

Artificial intelligence for stepwise diagnosis and monitoring of COVID-19.

BACKGROUND: Main challenges for COVID-19 include the lack of a rapid diagnostic test, a suitable tool to monitor and predict a patient's clinical course and an efficient way for data sharing among multicenters. We thus developed a novel artificial intelligence system based on deep learning (DL) and federated learning (FL) for the diagnosis, monitoring, and prediction of a patient's clinical course. METHODS: CT imaging derived from 6 different multicenter cohorts were used for stepwise diagnostic algorithm to diagnose COVID-19, with or without clinical data. Patients with more than 3 consecutive CT images were trained for the monitoring algorithm. FL has been applied for decentralized refinement of independently built DL models. RESULTS: A total of 1,552,988 CT slices from 4804 patients were used. The model can diagnose COVID-19 based on CT alone with the AUC being 0.98 (95% CI 0.97-0.99), and outperforms the radiologist's assessment. We have also successfully tested the incorporation of the DL diagnostic model with the FL framework. Its auto-segmentation analyses co-related well with those by radiologists and achieved a high Dice's coefficient of 0.77. It can produce a predictive curve of a patient's clinical course if serial CT assessments are available. INTERPRETATION: The system has high consistency in diagnosing COVID-19 based on CT, with or without clinical data. Alternatively, it can be implemented on a FL platform, which would potentially encourage the data sharing in the future. It also can produce an objective predictive curve of a patient's clinical course for visualization. KEY POINTS: • CoviDet could diagnose COVID-19 based on chest CT with high consistency; this outperformed the radiologist's assessment. Its auto-segmentation analyses co-related well with those by radiologists and could potentially monitor and predict a patient's clinical course if serial CT assessments are available. It can be integrated into the federated learning framework. • CoviDet can be used as an adjunct to aid clinicians with the CT diagnosis of COVID-19 and can potentially be used for disease monitoring; federated learning can potentially open opportunities for global collaboration.