Landscape of 2D Deep Learning Segmentation Networks Applied to CT Scan from Lung Cancer Patients: A Systematic Review

Deleted Journal, Journal Year: 2025, Volume and Issue: unknown

Published: March 4, 2025

The increasing rates of lung cancer emphasize the need for early detection through computed tomography (CT) scans, enhanced by deep learning (DL) to improve diagnosis, treatment, and patient survival. This review examines current prospective applications 2D- DL networks in CT segmentation, summarizing research, highlighting essential concepts gaps; Methods: Following Preferred Reporting Items Systematic Reviews Meta-Analysis guidelines, a systematic search peer-reviewed studies from 01/2020 12/2024 on data-driven population segmentation using structured data was conducted across databases like Google Scholar, PubMed, Science Direct, IEEE (Institute Electrical Electronics Engineers) ACM (Association Computing Machinery) library. 124 met inclusion criteria were analyzed. LIDC-LIDR dataset most frequently used; finding particularly relies supervised with labeled data. UNet model its variants used models medical image achieving Dice Similarity Coefficients (DSC) up 0.9999. reviewed primarily exhibit significant gaps addressing class imbalances (67%), underuse cross-validation (21%), poor stability evaluations (3%). Additionally, 88% failed address missing data, generalizability concerns only discussed 34% cases. emphasizes importance Convolutional Neural Networks, UNet, analysis advocates combined 2D/3D modeling approach. It also highlights larger, diverse datasets exploration semi-supervised unsupervised enhance automated diagnosis detection.

Language: Английский

Development and validation of machine learning models based on molecular features for estimating the probability of multiple primary lung carcinoma versus intrapulmonary metastasis in patients presenting multiple non-small cell lung cancers

Translational Lung Cancer Research, Journal Year: 2025, Volume and Issue: 14(4), P. 1118 - 1137

Published: April 1, 2025

Discrimination of multiple non-small cell lung cancers (NSCLCs) as primary (MPLCs) or intrapulmonary metastases (IPMs) is critical but remains challenging. The aim this study to develop and validate the machine learning (ML) models based on molecular features for estimating probability MPLC IPM patients presenting NSCLCs. A total 72 NSCLCs with 157 surgical resection tumor lesions from January 2012 2018 at two institutions were included developing testing models. Specifically, 46 103 tumors which defined definitive according International Association Study Lung Cancer (IASLC) criteria used They spilt into training validation sets using stratified random sampling five-fold cross-validation. developed tested in other 26 whose undetermined by traditional methods. Whole-exome sequencing (WES) was performed all samples. Four calculated characterize relatedness served model inputs, including genetic divergence, shared mutation number, Pearson correlation coefficient early number. Decision trees (DT), forests (RF), gradient boosting decision (GBDT) employed, performance assessed areas under curve (AUCs), accuracy, precision, recall, F1 score set. Disease-free survival (DFS) evaluate test cohort. Clinical characteristics then compared between populations. All four showed significant differences development That is, exhibited higher lower number than (P<0.001). DT model, RF GBDT these factors achieved a mean AUC 0.94 [standard deviation (SD) 0.09], 1.00 (SD 0.00) set, respectively. discriminated (n=15) (n=11) consistently. identified ML had significantly prolonged DFS [hazard ratio =0.21; 95% confidence interval (CI): 0.04-1.0; P=0.04] that IPM. relative prevalence family history first-degree relatives cancer, more half reported cancer. EGFR most common mutated driver both effectively distcriminate NSCLCs, improve accuracy diagnosis assist clinical decision-making, particularly challenging cases.

Language: Английский