Interactive exploration of a global clinical network from a large breast cancer cohort

Abstract

Despite unprecedented amount of information now available in medical records, health data remain underexploited due to their heterogeneity and complexity. Simple charts and hypothesis-driven statistics can no longer apprehend the content of information-rich clinical data. There is, therefore, a clear need for powerful interactive visualization tools enabling medical practitioners to perceive the patterns and insights gained by state-of-the-art machine learning algorithms. Here, we report an interactive graphical interface for use as the front end of a machine learning causal inference server (MIIC), to facilitate the visualization and comprehension by clinicians of relationships between clinically relevant variables. The widespread use of such tools, facilitating the interactive exploration of datasets, is crucial both for data visualization and for the generation of research hypotheses. We demonstrate the utility of the MIIC interactive interface, by exploring the clinical network of a large cohort of breast cancer patients treated with neoadjuvant chemotherapy (NAC). This example highlights, in particular, the direct and indirect links between post-NAC clinical responses and patient survival. The MIIC interactive graphical interface has the potential to help clinicians identify actionable nodes and edges in clinical networks, thereby ultimately improving the patient care pathway.

CausalXtract: a flexible pipeline to extract causal effects from live-cell time-lapse imaging data

“…1c, is adapted from the causal discovery method, MIIC [4][5][6] , which learns contemporaneous causal networks (i.e. when temporal information is not available) for a broad range of biological or biomedical data, from single-cell transcriptomic and genomic alteration data 4,7 to medical records of patients 5,6,8 . Live-cell time-lapse imaging data contain, however, information about cellular dynamics, which can in principle facilitate the discovery of novel cause-effect functional processes, based on the assumption that future events cannot cause past ones.…”

Learning interpretable causal networks from very large datasets, application to 400,000 medical records of breast cancer patients

“…In particular, iMIIC automatically adjusts for measured confounders (in the form of indirect contributions) and distinguishes genuine causes from putative and latent causal effects by either ruling out or highlighting the effect of unmeasured confounders for each causal edge ( Figures 2 and S1 ). While iMIIC is not immune to possible data collection and selection biases, which can affect observational data, it is based on a robust information theoretic framework, making it particularly reliable to interpret challenging types of data, such as heterogeneous data including combination of continuous and categorical variables integrated from different sources (e.g., clinical, personal, socio-economic data, as demonstrated here and on much smaller datasets in earlier studies 10 , 48 ) or different experimental techniques (e.g., single cell transcriptomics 8 , 44 , 45 , 46 and imaging data 10 , 47 ). In principle, iMIIC could be applied to a wide range of other domains to uncover causal relations and quantify indirect contributions when only observational data is available.…”

Trend of Women’s Health Research in Korea, 2012–2020: Topic and Text Network Analysis

“… 13 The dominance of breast cancer research in the field of women’s health research appears to be attributable to several factors: public awareness of its higher incidence, the development of innovative anti-cancer drugs, a well-established research network, diverse funding resources, good accessibility to biomaterials, and the active participation of patients in clinical trials. 11 30 31 32 33 Given the high burden of musculoskeletal disorders, diabetes, urogenital disorders, and blood and endocrine diseases in Korean women, more attention is required to diverse health issues across the life course of women. 34 The lack of diversity in women’s health studies was also illustrated shown by the current result of text analysis.…”