Machine Learning Methods for Survival Analysis with Clinical and Transcriptomics Data of Breast Cancer

Abstract

Breast cancer is one of the most common cancers in women worldwide, which causes an enormous number of deaths annually. However, early diagnosis of breast cancer can improve survival outcomes enabling simpler and more costeffective treatments. The recent increase in data availability provides unprecedented opportunities to apply data-driven and machine learning methods to identify earlydetection prognostic factors capable of predicting the expected survival and potential sensitivity to treatment of patients, with the final aim of enhancing clinical outcomes.

The Effects of Omega-3 Polyunsaturated Fatty Acids on Breast Cancer as a Preventive Measure or as an Adjunct to Conventional Treatments

“…The available treatments consist of partial or total mastectomy, chemotherapy, hormone therapy, immunotherapy, and/or radiotherapy [ 10 , 11 ]. When treated properly and in a timely manner, BC presents a good prognosis, in spite of chemo- or radiotherapy-related side effects that impact patients’ quality of life [ 12 , 13 ] and often lead to an increased treatment discontinuation rate [ 3 , 10 , 13 ].…”

Editorial: Artificial intelligence for data discovery and reuse in endocrinology and metabolism

“…Overall, they emphasise that the combination of multiple modalities is more effective than using only one modality in isolation, showing a significant increase in predictive performance. They also address the issue of small sample size, a common drawback of ML studies in omics, where obtaining matched samples across more than one modality remains a challenging task in of time and costs involved, with the associated challenges in using deep learning approaches (3,4).…”

Cross-attention enables deep learning on limited omics-imaging-clinical data of 130 lung cancer patients

“…Such techniques are best suited for high-dimensional data because of their ability to perform survival analysis using both statistical and machine-learning methods. 27 , 28 , 29 …”