Development and Application of a Novel Machine Learning Model Predicting Pancreatic Cancer-Specific Mortality DOI Open Access

Yongji Sun,

Sien Hu, Xiawei Li

и другие.

Cureus, Год журнала: 2024, Номер unknown

Опубликована: Март 29, 2024

Precise prognostication is vital for guiding treatment decisions in people diagnosed with pancreatic cancer. Existing models depend on predetermined variables, constraining their effectiveness. Our objective was to explore a novel machine learning approach enhance prognostic model predicting cancer-specific mortality and, subsequently, assess its performance against Cox regression models. Datasets were retrospectively collected and analyzed 9,752 patients cancer surgery performed. The primary outcomes the of carcinoma at one year, three years, five years. Model discrimination assessed using concordance index (C-index), calibration Brier scores. Survival Quilts compared clinical use, decision curve analysis done. demonstrated robust one-year (C-index 0.729), three-year 0.693), five-year 0.672) mortality. In comparison models, exhibited higher C-index up 32 months but displayed inferior after 33 months. A subgroup conducted, revealing that within subset individuals without metastasis, showcased significant advantage over cohort metastatic cancer, outperformed before 24 weaker 25 This study has developed validated learning-based predict outperforms model.

Язык: Английский

Integrated multiomics analysis and machine learning refine molecular subtypes and prognosis for muscle-invasive urothelial cancer DOI Creative Commons
Guangdi Chu, Xiaoyu Ji, Yonghua Wang

и другие.

Molecular Therapy — Nucleic Acids, Год журнала: 2023, Номер 33, С. 110 - 126

Опубликована: Июнь 5, 2023

Muscle-invasive urothelial cancer (MUC), characterized by high aggressiveness and significant heterogeneity, is currently lacking highly precise individualized treatment options. We used a computational pipeline to synthesize multiomics data from MUC patients using 10 clustering algorithms, which were then combined with machine learning algorithms identify molecular subgroups of resolution develop robust consensus learning-driven signature (CMLS). Through clustering, we identified three subtypes (CSs) that are related prognosis, CS2 exhibiting the most favorable prognostic outcome. Subsequent screening enabled identification 12 hub genes constitute CMLS predictive power for prognosis. The low-CMLS group exhibited more prognosis greater responsiveness immunotherapy was likely exhibit "hot tumor" phenotype. high-CMLS had poor lower likelihood benefitting immunotherapy, but dasatinib romidepsin may serve as promising treatments them. Comprehensive analysis can offer important insights further refine classification MUC. Identification represents valuable tool early prediction patient potential candidates benefit broad implications clinical practice.

Язык: Английский

Процитировано

42
Identifying squalene epoxidase as a metabolic vulnerability in high‐risk osteosarcoma using an artificial intelligence‐derived prognostic index DOI Creative Commons
Yongjie Wang, Xiaolong Ma, Enjie Xu

и другие.

Clinical and Translational Medicine, Год журнала: 2024, Номер 14(2)

Опубликована: Фев. 1, 2024

Osteosarcoma (OSA) presents a clinical challenge and has low 5-year survival rate. Currently, the lack of advanced stratification models makes personalized therapy difficult. This study aims to identify novel biomarkers stratify high-risk OSA patients guide treatment.

Язык: Английский

Процитировано

6
Robust machine−learning based prognostic index using cytotoxic T lymphocyte evasion genes highlights potential therapeutic targets in colorectal cancer DOI Creative Commons
Xu Wang,

Shixin Chan,

Jiajie Chen

и другие.

Cancer Cell International, Год журнала: 2024, Номер 24(1)

Опубликована: Янв. 31, 2024

A minute fraction of patients stands to derive substantial benefits from immunotherapy, primarily attributable immune evasion. Our objective was formulate a predictive signature rooted in genes associated with cytotoxic T lymphocyte evasion (CERGs), the aim predicting outcomes and discerning immunotherapeutic response colorectal cancer (CRC).

Язык: Английский

Процитировано

5
Identification macrophage signatures in prostate cancer by single-cell sequencing and machine learning DOI Creative Commons

Zhen Kang,

Yuxuan Zhao,

Ren Qiu

и другие.

Cancer Immunology Immunotherapy, Год журнала: 2024, Номер 73(3)

Опубликована: Фев. 13, 2024

Abstract Background The tumor microenvironment (TME) encompasses a variety of cells that influence immune responses and growth, with tumor-associated macrophages (TAM) being crucial component the TME. TAM can guide prostate cancer in different directions response to various external stimuli. Methods First, we downloaded single-cell sequencing data second-generation from multiple public databases. From these data, identified characteristic genes associated clusters. We then employed machine learning techniques select most accurate gene set developed TAM-related risk label for cancer. analyzed tumor-relatedness groups within population. Finally, validated accuracy prognostic using qPCR, WB assays, among other methods. Results In this study, TAM_2 cell cluster has been as promoting progression cancer, possibly representing M2 macrophages. 9 feature selected through ten methods demonstrated their effectiveness predicting patients. Additionally, have linked clinical pathological characteristics, allowing us construct nomogram. This nomogram provides practitioners quantitative tool assessing prognosis Conclusion study potential relationship between PCa established model. It holds promise valuable management treatment

Язык: Английский

Процитировано

5
Integrating PANoptosis insights to enhance breast cancer prognosis and therapeutic decision-making DOI Creative Commons
Shu Wang, Zhuolin Li, Jing Hou

и другие.

Frontiers in Immunology, Год журнала: 2024, Номер 15

Опубликована: Март 5, 2024

Background Despite advancements, breast cancer outcomes remain stagnant, highlighting the need for precise biomarkers in precision medicine. Traditional TNM staging is insufficient identifying patients who will respond well to treatment. Methods Our study involved over 6,900 from 14 datasets, including in-house clinical data and single-cell 8 (37,451 cells). We integrated 10 machine learning algorithms 55 combinations analyzed 100 existing signatures. IHC assays were conducted validation, potential immunotherapies chemotherapies explored. Results pinpointed six stable Panoptosis-related genes multi-center cohorts, leading a robust Panoptosis-model. This model outperformed molecular features predicting recurrence mortality risks, with high-risk showing worse outcomes. validation 30 confirmed our findings, indicating model’s broader applicability. Additionally, suggested that low-risk benefit more immunotherapy, while are sensitive specific like BI-2536 ispinesib. Conclusion The Panoptosis-model represents major advancement prognosis treatment personalization, offering significant insights effectively managing wide range of patients.

Язык: Английский

Процитировано

5
Integration analysis of cell division cycle-associated family genes revealed potential mechanisms of gliomagenesis and constructed an artificial intelligence-driven prognostic signature DOI

Kai Yu,

Qi Tian,

Shi Feng

и другие.

Cellular Signalling, Год журнала: 2024, Номер 119, С. 111168 - 111168

Опубликована: Апрель 9, 2024

Язык: Английский

Процитировано

5
Unveiling the Potential of Migrasomes: A Machine-Learning-Driven Signature for Diagnosing Acute Myocardial Infarction DOI Creative Commons
Yihao Zhu, Yuxi Chen,

Jiajin Xu

и другие.

Biomedicines, Год журнала: 2024, Номер 12(7), С. 1626 - 1626

Опубликована: Июль 22, 2024

Recent studies have demonstrated that the migrasome, a newly functional extracellular vesicle, is potentially significant in occurrence, progression, and diagnosis of cardiovascular diseases. Nonetheless, its diagnostic significance biological mechanism acute myocardial infarction (AMI) yet to be fully explored.

Язык: Английский

Процитировано

5
Epigenetic profiling for prognostic stratification and personalized therapy in breast cancer DOI Creative Commons

Xiao Guo,

Chuanbo Feng,

Jiaying Xing

и другие.

Frontiers in Immunology, Год журнала: 2025, Номер 15

Опубликована: Янв. 14, 2025

Background The rising incidence of breast cancer and its heterogeneity necessitate precise tools for predicting patient prognosis tailoring personalized treatments. Epigenetic changes play a critical role in progression therapy responses, providing foundation prognostic model development. Methods We developed the Machine Learning-derived Model (MLEM) to identify epigenetic gene patterns cancer. Using multi-cohort transcriptomic datasets, MLEM was constructed with rigorous machine learning techniques validated across independent datasets. model’s performance further corroborated through immunohistochemical validation on clinical samples. Results effectively stratified patients into high- low-risk groups. Low-MLEM exhibited improved prognosis, characterized by enhanced immune cell infiltration higher responsiveness immunotherapy. High-MLEM showed poorer but were more responsive chemotherapy, vincristine identified as promising therapeutic option. demonstrated robust Conclusion is powerful tool outcomes By integrating insights learning, this has potential improve decision-making optimize strategies patients.

Язык: Английский

Процитировано

0
Uncovering the Potential of Pathomics: Prognostic Prediction and Mechanistic Investigation of Pancreatic Cancer DOI
Rixin Su, Xiaohong Zhao, Fabiao Zhang

и другие.

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
Integrative machine learning model of RNA modifications predict prognosis and treatment response in patients with breast cancer DOI Creative Commons
Tao Wang, Shu Wang, Zhuolin Li

и другие.

Cancer Cell International, Год журнала: 2025, Номер 25(1)

Опубликована: Фев. 13, 2025

Breast cancer, a highly heterogeneous and complex disease, remains the leading cause of cancer-related death among women worldwide. Despite advances in treatment modalities, effective prognostic models therapeutic strategies are still urgently needed. We retrospectively analyzed 15 independent breast cancer cohorts to explore role RNA modifications prognosis patients with cancer. By integrating nine types modifications, we developed comprehensive machine learning-based modification signature (CMRS). Furthermore, single-cell sequencing data were understand biological mechanisms underlying CMRS. In addition, immune infiltration levels evaluated via six different algorithms, checkpoint inhibitor responsiveness was predicted. Moreover, response high-CMIS chemotherapy predicted multiple datasets. Finally, immunohistochemistry performed on tissue samples from validate protein expression levels. Our analysis revealed five key modification-related genes (ENO1, ARAF, WT1, GADD45A, BIRC3) associated prognosis. The CMRS model demonstrated high predictive accuracy across significantly correlated patient survival outcomes. Multiomics that increased tumor mutational burden distinct signatures, particularly pathways related TP53, MYC, cell proliferation. Single-cell highlighted involvement epithelial cells MYC signaling activity. Cell‒cell communication reduced interaction strength hig patients, indicating poor low presented improved inhibitors, whereas identified as potential candidates for panobinostat vincristine. study elucidates significant treatment. serves sensitive biomarker predicting responsiveness, offering new avenue personalized therapy

Язык: Английский

Процитировано

0