Predicting gastric cancer response to anti-HER2 therapy or anti-HER2 combined immunotherapy based on multi-modal data

Signal Transduction and Targeted Therapy, Journal Year: 2024, Volume and Issue: 9(1)

Published: Aug. 26, 2024

The sole use of single modality data often fails to capture the complex heterogeneity among patients, including variability in resistance anti-HER2 therapy and outcomes combined treatment regimens, for HER2-positive gastric cancer (GC). This deficit has not been fully considered many studies. Furthermore, application artificial intelligence predicting response, particularly diseases such as GC, is still its infancy. Therefore, this study aimed a comprehensive analytic approach accurately predict responses or immunotherapy patients with GC. We collected multi-modal data, comprising radiology, pathology, clinical information from cohort 429 patients: 310 treated 119 combination anti-PD-1/PD-L1 inhibitors immunotherapy. introduced deep learning model, called Multi-Modal model (MuMo), that integrates these make precise response predictions. MuMo achieved an area under curve score 0.821 0.914 Moreover, classified low-risk by exhibited significantly prolonged progression-free survival overall (log-rank test, P < 0.05). These findings only highlight significance analysis enhancing evaluation personalized medicine cancer, but also potential value our model.

Language: Английский

---

Utilizing machine learning for predicting drug release from polymeric drug delivery systems

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 188, P. 109756 - 109756

Published: Feb. 19, 2025

Language: Английский

---

Artificial intelligence and pathology: From principles to practice and future applications in histomorphology and molecular profiling

Seminars in Cancer Biology, Journal Year: 2021, Volume and Issue: 84, P. 129 - 143

Published: Feb. 22, 2021

Language: Английский

---

A cross-study analysis of drug response prediction in cancer cell lines

Briefings in Bioinformatics, Journal Year: 2021, Volume and Issue: 23(1)

Published: Sept. 8, 2021

Abstract To enable personalized cancer treatment, machine learning models have been developed to predict drug response as a function of tumor and features. However, most algorithm development efforts relied on cross-validation within single study assess model accuracy. While an essential first step, biological data set typically provides overly optimistic estimate the prediction performance independent test sets. provide more rigorous assessment generalizability between different studies, we use analyze five publicly available cell line-based sets: National Cancer Institute 60, ancer Therapeutics Response Portal (CTRP), Genomics Drug Sensitivity in Cancer, Cell Line Encyclopedia Genentech Screening Initiative (gCSI). Based observed experimental variability across explore estimates upper bounds. We report results variety models, with multitasking deep neural network achieving best cross-study generalizability. By multiple measures, trained CTRP yield accurate predictions remaining testing data, gCSI is predictable among line sets included this study. With these experiments further simulations partial two lessons emerge: (1) differences viability assays can limit studies (2) diversity, than crucial for raising preclinical screening.

Language: Английский

---

Chemometric approach in environmental pollution analysis: A critical review

Journal of Environmental Management, Journal Year: 2022, Volume and Issue: 309, P. 114653 - 114653

Published: Feb. 14, 2022

Language: Английский

---