Fig. 5: Interpreted pathological features. | npj Precision Oncology

Fig. 5: Interpreted pathological features.

From: Accurate prediction of disease-free and overall survival in non-small cell lung cancer using patient-level multimodal weakly supervised learning

Fig. 5: Interpreted pathological features.The alternative text for this image may have been generated using AI.

a The comparison of the number of hotspots for 5-year progression and 5-year death at patient and slide levels. P > D: the number of hotspots for 5-year progression exceeded that for 5-year death (patients: 101/125 (80.8%); slides: 369/536 (68.8%)); P = D: the number of hotspots for 5-year progression was comparable to that for 5-year death (patients: 17/125 (13.6%); slides: 114/536 (21.3%)); P < D: the number of hotspots for 5-year progression was fewer than that for 5-year death (patients: 7/125 (5.6%); slides: 53/536 (9.9%)). b The composition of histological types of NSCLC patients in the test set. c The distribution of hotspots on WSIs for patients with SCC, along with the enlargement of example regions. Red arrow: mitotic figures; Green arrow: enlarged and bizarre nuclei. d The actual and predicted results in NMA subtypes. Reality: the number of patients with the NMA subtype; prediction: the number of patients with risk hotspots within the subtype. The distribution of hotspots on WSIs for patients with MPA (e), SPA (f), LPA (g), along with the enlargement of example regions. h The distribution of hotspots on WSIs for patients with APA. The enlarged regions display the features of small glands(upside) and big glands(underside). i The stromal regions identified as risk hotspots. j The distribution of hotspots on WSIs for patients with PPA, along with the enlargement of example regions. SCC squamous cell carcinomas, NMA non-mucinous adenocarcinoma, SPA solid adenocarcinoma, MPA micropapillary adenocarcinoma, LPA lepidic adenocarcinoma, APA acinar adenocarcinoma, PPA papillary adenocarcinoma.

Back to article page