Prediction of lung metastases in thyroid cancer using machine learning based on <scp>SEER</scp> database

Liu, Wenfei; Wang, Shoufei; Ye, Ziheng; Xu, Peipei; Xia, Xiaotian; Guo, Minggao

doi:10.1002/cam4.4617

Cited by 44 publications

(31 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Contemporarily, machine learning has become a powerful helper in the medical field, showing great potential in disease diagnosis, 57 , 58 medical image processing, 59 , 60 etc. In this present study, four machine‐learning algorithms including SVM, RF, XGBoost, and LightGBM were applied to predict the OS of OC patients.…”

Section: Discussionmentioning

confidence: 99%

Peripheral and tumor‐infiltrating immune cells are correlated with patient outcomes in ovarian cancer

et al. 2023

View full text Add to dashboard Cite

Objective At present, there is still a lack of reliable biomarkers for ovarian cancer (OC) to guide prognosis prediction and accurately evaluate the dominant population of immunotherapy. In recent years, the relationship between peripheral blood markers and tumor‐infiltrating immune cells (TICs) with cancer has attracted much attention. However, the relationship between the survival of OC patients and intratumoral‐ or extratumoral‐associated immune cells remains controversial. Methods In this study, four machine‐learning algorithms were used to predict overall survival in OC patients based on peripheral blood indicators. To further screen out immune‐related gene and molecular targets, we systematically explored the correlation between TICs and OC patient survival based on The Cancer Genome Atlas database. Using the TICs score method, patients were divided into a low immune infiltrating cell group and a high immune infiltrating cell group. Results The results showed that there was a significant statistical significance between the peripheral blood indicators and the survival prognosis of OC patients. Survival analysis showed that TICs play a crucial role in the survival of OC patients. Four core genes, CXCL9, CD79A, MS4A1, and MZB1, were identified by cross‐PPI and COX regression analysis. Further analysis found that these genes were significantly associated with both TICs and survival in OC patients. Conclusions These results suggest that both peripheral blood markers and TICs can be used as prognostic predictors in patients with OC, and CXCL9, CD79A, MS4A1, and MZB1 may be potential therapeutic targets for OC immunotherapy.

show abstract

Section: Discussionmentioning

confidence: 99%

Peripheral and tumor‐infiltrating immune cells are correlated with patient outcomes in ovarian cancer

et al. 2023

View full text Add to dashboard Cite

show abstract

“…In the selected best models, the evaluation indices, including AUC (0.99) and Brier score (0.016), were found to be excellent. Age, T-stage, N-stage, and histological type were identi ed as in uencing factors for lung metastasis occurrence [27]. Additionally, a large dataset of 207,137 prostate cancer patients was used to predict bone metastasis using machine learning techniques.…”

Section: Discussionmentioning

confidence: 99%

Machine Learning Model for Predicting Brain Metastases in Lung Adenocarcinomas

Tao,

Wang

et al. 2023

Preprint

View full text Add to dashboard Cite

Background Precisely predicting the onset of brain metastasis in patients with lung adenocarcinoma is critical for making informed clinical treatment decisions. After brain metastasis occurs, the survival period of patients with lung adenocarcinoma is substantially reduced, and a dearth of models hinders accurate prediction of its onset in affected patients. In this study, we compared the performance of five models and identified the random forest model as the most effective for predicting brain metastasis in patients with lung adenocarcinoma. Methods This study enrolled patients diagnosed with lung adenocarcinoma between 2000 to 2018, sourced from the Surveillance, Epidemiology, and End Results (SEER) database. Correlation between variables was observed using a heatmap and the model's discriminative ability was assessed by calculating the Area Under the Receiver Operating Characteristic curve (AUROC). In addition, features were ranked according to their importance. Furthermore, the precision of the model was assessed by means of calibration curves, and a decision curve analysis was performed to evaluate its clinical usefulness. Results A total of 60,805 patients were included in this study. The heatmap analysis provided a preliminary assessment of the correlation between different feature variables, and all variables showed differences between cases with and without brain metastasis after standardization. The random forest model exhibited better predictive performance with an accuracy of 0.919 (95% CI: 0.915–0.926) and an AUROC of 0.92 (95% CI: 0.913–0.924). Regarding feature importance ranking, the most relevant features were radiation therapy, survival time, tumor size, age, and bone metastasis. The calibration curve showed the highest degree of consistency between the predicted probabilities and actual probabilities in the random forest model. The decision curve analysis revealed a considerable enhancement in net benefit for the models containing 10 features, 17 features, and all features, as compared to the basic model. Conclusions We developed predictive model using machine learning to predict brain metastasis in patients with lung adenocarcinoma, utilizing only commonly available clinical variables. The model showed good discrimination in predicting the occurrence of brain metastasis. It may contribute to clinical decision-making and treatment strategies.

show abstract

“…To enable early identification of patients with DM risks, several clinic-pathological features have been proposed. Older age [3], male sex [4], and pathologic factors like T stage [5], lymph node metastasis (LNM) [6], and extrathyroidal extension (ETE) [7] were reported to predict DM. Besides, genetic alterations among candidate genes were also postulated to improve the risk stratification of TC in recent years.…”

Section: Discussionmentioning

confidence: 99%

Risk-stratified Distant Metastatic Thyroid Cancer with Clinicopathological Factors and BRAF/TERT Promoter Mutations

Cheng,

Zhou,

et al. 2023

Exp Clin Endocrinol Diabetes

View full text Add to dashboard Cite

Objective To assess the prognostic value of clinicopathological factors as well as BRAF and TERT promoter mutations in predicting distant metastasis in patients with papillary thyroid cancer. Design The status of BRAF and TERTp mutations were available in 1,208 thyroid cancer patients who received thyroidectomy at Jiangyuan Hospital Affiliated to Jiangsu Institute of Nuclear Medicine from January 2008 to December 2021. Based on inclusion criteria, 99 distant metastasis thyroid cancers (DM-TCs) and 1055 patients without DM (Non-DM-TCs) were retrospectively reviewed. Results After univariate and multivariate analyses, a risk model was established for DM prediction based on factors: T3/T4 stage, lymph node metastasis (LNM) number over 5, and BRAF/TERT mutations (TLBT). It was defined based on the number of TLBT factors: low risk (no risk factor, n=896), intermediate risk (1 risk factor, n=199), and high risk (≥2 risk factors, n=59). Notably, compared with patients with low and intermediate risks, patients assigned to high TLBT risk have a shorter time of DM disease-free survival. Except for gene mutation, other factors were also included in the 2015 American Thyroid Association (ATA) risk guideline. Comparing with the ATA risk category, this risk model showed a better performance in predicting DM-TCs. Conclusions This study proposes a TLBT risk classifier consisting of T3/T4 stages, LNM (n>5), and BRAF+TERTp mutations for predicting DM-TCs. TLBT risk stratification may help clinicians make personalized treatment management and follow-up strategies.

show abstract

Prediction of lung metastases in thyroid cancer using machine learning based on SEER database

Cited by 44 publications

References 50 publications

Peripheral and tumor‐infiltrating immune cells are correlated with patient outcomes in ovarian cancer

Peripheral and tumor‐infiltrating immune cells are correlated with patient outcomes in ovarian cancer

Machine Learning Model for Predicting Brain Metastases in Lung Adenocarcinomas

Risk-stratified Distant Metastatic Thyroid Cancer with Clinicopathological Factors and BRAF/TERT Promoter Mutations

Contact Info

Product

Resources

About