Predicting EGFR Mutation Status in Non-Small Cell Lung Cancer Using Artificial Intelligence: A Systematic Review and Meta-Analysis

A privacy-preserving platform oriented medical healthcare and its application in identifying patients with candidemia

Federated learning (FL) has emerged as a significant method for developing machine learning models across multiple devices without centralized data collection. Candidemia, a critical but rare disease in ICUs, poses challenges in early detection and treatment. The goal of this study is to develop a privacy-preserving federated learning framework for predicting candidemia in ICU patients. This approach aims to enhance the accuracy of antifungal drug prescriptions and patient outcomes. This study involved the creation of four predictive FL models for candidemia using data from ICU patients across three hospitals in China. The models were designed to prioritize patient privacy while aggregating learnings across different sites. A unique ensemble feature selection strategy was implemented, combining the strengths of XGBoost's feature importance and statistical test p values. This strategy aimed to optimize the selection of relevant features for accurate predictions. The federated learning models demonstrated significant improvements over locally trained models, with a 9% increase in the area under the curve (AUC) and a 24% rise in true positive ratio (TPR). Notably, the FL models excelled in the combined TPR + TNR metric, which is critical for feature selection in candidemia prediction. The ensemble feature selection method proved more efficient than previous approaches, achieving comparable performance. The study successfully developed a set of federated learning models that significantly enhance the prediction of candidemia in ICU patients. By leveraging a novel feature selection method and maintaining patient privacy, the models provide a robust framework for improved clinical decision-making in the treatment of candidemia.

Using Multi-phase CT Radiomics Features to Predict EGFR Mutation Status in Lung Adenocarcinoma Patients

RATIONALE AND OBJECTIVES

This study aimed to non-invasively predict epidermal growth factor receptor (EGFR) mutation status in patients with lung adenocarcinoma using multi-phase computed tomography (CT) radiomics features.

MATERIALS AND METHODS

A total of 424 patients with lung adenocarcinoma were recruited from two hospitals who underwent preoperative non-enhanced CT (NE-CT) and enhanced CT (including arterial phase CT [AP-CT], and venous phase CT [VP-CT]). Patients were divided into training (n = 297) and external validation (n = 127) cohorts according to hospital. Radiomics features were extracted from the NE-CT, AP-CT, and VP-CT images, respectively. The Wilcoxon test, correlation analysis, and simulated annealing were used for feature screening. A clinical model and eight radiomics models were established. Furthermore, a clinical-radiomics model was constructed by incorporating multi-phase CT features and clinical risk factors. Receiver operating characteristic curves were used to evaluate the predictive performance of the models.

RESULTS

The predictive performance of multi-phase CT radiomics model (AUC of 0.925 [95% CI, 0.879-0.971] in the validation cohort) was higher than that of NE-CT, AP-CT, VP-CT, and clinical models (AUCs of 0.860 [95% CI,0.794-0.927], 0.792 [95% CI, 0.713-0.871], 0.753 [95% CI, 0.669-0.838], and 0.706 [95% CI, 0.620-0.791] in the validation cohort, respectively) (all P < 0.05). The predictive performance of the clinical-radiomics model (AUC of 0.927 [95% CI, 0.882-0.971] in the validation cohort) was comparable to that of multi-phase CT radiomics model (P > 0.05).

CONCLUSION

Our multi-phase CT radiomics model showed good performance in identifying the EGFR mutation status in patients with lung adenocarcinoma, which may assist personalized treatment decisions.

Identification of biomarkers associated with ferroptosis in diabetic retinopathy based on WGCNA and machine learning

Objective: Diabetic retinopathy (DR) is a chronic progressive eye disease that affects millions of diabetic patients worldwide, and ferroptosis may contribute to the underlying mechanisms of DR. The main objective of this work is to explore key genes associated with ferroptosis in DR and to determine their feasibility as diagnostic markers. Methods: WGCNA identify the most relevant signature modules in DR. Machine learning methods were used to de-screen the feature genes. ssGSEA calculated the scoring of immune cells in the DR versus control samples and compared the associations with the core genes by Spearman correlation. Results: We identified 2,897 differential genes in DR versus normal samples. WGCNA found tan module to have the highest correlation with DR patients. Finally, 20 intersecting genes were obtained from differential genes, tan module and iron death genes, which were screened by LASSO and SVM-RFE method, and together identified 6 genes as potential diagnostic markers. qPCR verified the expression and ROC curves confirmed the diagnostic accuracy of the 6 genes. In addition, our ssGSEA scoring identified these 6 core genes as closely associated with immune infiltrating cells. Conclusion: In conclusion, we analyzed for the first time the potential link of iron death in the pathogenesis of DR. This has important implications for future studies of iron death-mediated pro-inflammatory immune mechanisms.

Predicting pressure injury risk in hospitalised patients using machine learning with electronic health records: a US multilevel cohort study

OBJECTIVE

To predict the risk of hospital-acquired pressure injury using machine learning compared with standard care.

DESIGN

We obtained electronic health records (EHRs) to structure a multilevel cohort of hospitalised patients at risk for pressure injury and then calibrate a machine learning model to predict future pressure injury risk. Optimisation methods combined with multilevel logistic regression were used to develop a predictive algorithm of patient-specific shifts in risk over time. Machine learning methods were tested, including random forests, to identify predictive features for the algorithm. We reported the results of the regression approach as well as the area under the receiver operating characteristics (ROC) curve for predictive models.

SETTING

Hospitalised inpatients.

PARTICIPANTS

EHRs of 35 001 hospitalisations over 5 years across 2 academic hospitals.

MAIN OUTCOME MEASURE

Longitudinal shifts in pressure injury risk.

RESULTS

The predictive algorithm with features generated by machine learning achieved significantly improved prediction of pressure injury risk (p<0.001) with an area under the ROC curve of 0.72; whereas standard care only achieved an area under the ROC curve of 0.52. At a specificity of 0.50, the predictive algorithm achieved a sensitivity of 0.75.

CONCLUSIONS

These data could help hospitals conserve resources within a critical period of patient vulnerability of hospital-acquired pressure injury which is not reimbursed by US Medicare; thus, conserving between 30 000 and 90 000 labour-hours per year in an average 500-bed hospital. Hospitals can use this predictive algorithm to initiate a quality improvement programme for pressure injury prevention and further customise the algorithm to patient-specific variation by facility.

CT-based radiomics for predicting Ki-67 expression in lung cancer: a systematic review and meta-analysis

Background

Radiomics, an emerging field, presents a promising avenue for the accurate prediction of biomarkers in different solid cancers. Lung cancer remains a significant global health challenge, contributing substantially to cancer-related mortality. Accurate assessment of Ki-67, a marker reflecting cellular proliferation, is crucial for evaluating tumor aggressiveness and treatment responsiveness, particularly in non-small cell lung cancer (NSCLC).

Methods

A systematic review and meta-analysis conducted following the preferred reporting items for systematic review and meta-analysis of diagnostic test accuracy studies (PRISMA-DTA) guidelines. Two authors independently conducted a literature search until September 23, 2023, in PubMed, Embase, and Web of Science. The focus was on identifying radiomics studies that predict Ki-67 expression in lung cancer. We evaluated quality using both Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) and the Radiomics Quality Score (RQS) tools. For statistical analysis in the meta-analysis, we used STATA 14.2 to assess sensitivity, specificity, heterogeneity, and diagnostic values.

Results

Ten retrospective studies were pooled in the meta-analysis. The findings demonstrated that the use of computed tomography (CT) scan-based radiomics for predicting Ki-67 expression in lung cancer exhibited encouraging diagnostic performance. Pooled sensitivity, specificity, and area under the curve (AUC) in training cohorts were 0.78, 0.81, and 0.85, respectively. In validation cohorts, these values were 0.78, 0.70, and 0.81. Quality assessment using QUADAS-2 and RQS indicated generally acceptable study quality. Heterogeneity in training cohorts, attributed to factors like contrast-enhanced CT scans and specific Ki-67 thresholds, was observed. Notably, publication bias was detected in the training cohort, indicating that positive results are more likely to be published than non-significant or negative results. Thus, journals are encouraged to publish negative results as well.

Conclusion

In summary, CT-based radiomics exhibit promise in predicting Ki-67 expression in lung cancer. While the results suggest potential clinical utility, additional research efforts should concentrate on enhancing diagnostic accuracy. This could pave the way for the integration of radiomics methods as a less invasive alternative to current procedures like biopsy and surgery in the assessment of Ki-67 expression.

The effect of data resampling methods in radiomics

Radiomic datasets can be class-imbalanced, for instance, when the prevalence of diseases varies notably, meaning that the number of positive samples is much smaller than that of negative samples. In these cases, the majority class may dominate the model's training and thus negatively affect the model's predictive performance, leading to bias. Therefore, resampling methods are often utilized to class-balance the data. However, several resampling methods exist, and neither their relative predictive performance nor their impact on feature selection has been systematically analyzed. In this study, we aimed to measure the impact of nine resampling methods on radiomic models utilizing a set of fifteen publicly available datasets regarding their predictive performance. Furthermore, we evaluated the agreement and similarity of the set of selected features. Our results show that applying resampling methods did not improve the predictive performance on average. On specific datasets, slight improvements in predictive performance (+ 0.015 in AUC) could be seen. A considerable disagreement on the set of selected features was seen (only 28.7% of features agreed), which strongly impedes feature interpretability. However, selected features are similar when considering their correlation (82.9% of features correlated on average).

Predictive value of 18F-FDG PET/CT radiomics for EGFR mutation status in non-small cell lung cancer: a systematic review and meta-analysis

Objective

This study aimed to evaluate the value of 18F-FDG PET/CT radiomics in predicting EGFR gene mutations in non-small cell lung cancer by meta-analysis.

Methods

The PubMed, Embase, Cochrane Library, Web of Science, and CNKI databases were searched from the earliest available date to June 30, 2023. The meta-analysis was performed using the Stata 15.0 software. The methodological quality and risk of bias of included studies were assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 and Radiomics Quality Score criteria. The possible causes of heterogeneity were analyzed by meta-regression.

Results

A total of 17 studies involving 3763 non-small cell lung cancer patients were finally included. We analyzed 17 training cohorts and 10 validation cohorts independently. Within the training cohort, the application of 18F-FDG PET/CT radiomics in predicting EGFR mutations in NSCLC demonstrated a sensitivity of 0.76 (95% CI: 0.70-0.81) and a specificity of 0.78 (95% CI: 0.74-0.82), accompanied by a positive likelihood ratio of 3.5 (95% CI:3.0-4.2), a negative likelihood ratio of 0.31 (95% CI: 0.24-0.39), a diagnostic odds ratio of 11.0 (95% CI: 8.0-16.0), and an area under the curve (AUC) of 0.84 (95% CI: 0.80-0.87). In the validation cohort, the values included a sensitivity of 0.76 (95% CI: 0.67-0.83), a specificity of 0.75 (95% CI: 0.68-0.80), a positive likelihood ratio of 3.0 (95% CI:2.4-3.8), a negative likelihood ratio of 0.32 (95% CI: 0.24-0.44), a diagnostic odds ratio of 9 (95% CI: 6-15), and an AUC of 0.82 (95% CI: 0.78-0.85). The average Radiomics Quality Score (RQS) across studies was 10.47 ± 4.72. Meta-regression analysis identifies the application of deep learning and regions as sources of heterogeneity.

Conclusion

18F-FDG PET/CT radiomics may be useful in predicting mutation status of the EGFR gene in non-small cell lung cancer.

Systematic review registration

https://www.crd.york.ac.uk/PROSPERO, identifier CRD42022385364.

Blood FOLR3 methylation dysregulations and heterogeneity in non-small lung cancer highlight its strong associations with lung squamous carcinoma

BACKGROUND

Non-small cell lung cancer (NSCLC) accounts for the vast majority of lung cancers. Early detection is crucial to reduce lung cancer-related mortality. Aberrant DNA methylation occurs early during carcinogenesis and can be detected in blood. It is essential to investigate the dysregulated blood methylation markers for early diagnosis of NSCLC.

METHODS

NSCLC-associated methylation gene folate receptor gamma (FOLR3) was selected from an Illumina 850K array analysis of peripheral blood samples. Mass spectrometry was used for validation in two independent case-control studies (validation I: n = 2548; validation II: n = 3866). Patients with lung squamous carcinoma (LUSC) or lung adenocarcinoma (LUAD), normal controls (NCs) and benign pulmonary nodule (BPN) cases were included. FOLR3 methylations were compared among different populations. Their associations with NSCLC clinical features were investigated. Receiver operating characteristic analyses, Kruskal-Wallis test, Wilcoxon test, logistics regression analysis and nomogram analysis were performed.

RESULTS

Two CpG sites (CpG_1 and CpG_2) of FOLR3 was significantly lower methylated in NSCLC patients than NCs in the discovery round. In the two validations, both LUSC and LUAD patients presented significant FOLR3 hypomethylations. LUSC patients were highlighted to have significantly lower methylation levels of CpG_1 and CpG_2 than BPN cases and LUAD patients. Both in the two validations, CpG_1 methylation and CpG_2 methylation could discriminate LUSC from NCs well, with areas under the curve (AUCs) of 0.818 and 0.832 in validation I, and 0.789 and 0.780 in validation II. They could also differentiate LUAD from NCs, but with lower efficiency. CpG_1 and CpG_2 methylations could also discriminate LUSC from BPNs well individually in the two validations. With the combined dataset of two validations, the independent associations of age, gender, and FOLR3 methylation with LUSC and LUAD risk were shown and the age-gender-CpG_1 signature could discriminate LUSC and LUAD from NCs and BPNs, with higher efficiency for LUSC.

CONCLUSIONS

Blood-based FOLR3 hypomethylation was shown in LUSC and LUAD. FOLR3 methylation heterogeneity between LUSC and LUAD highlighted its stronger associations with LUSC. FOLR3 methylation and the age-gender-CpG_1 signature might be novel diagnostic markers for the early detection of NSCLC, especially for LUSC.

Applying Sequential Pattern Mining to Investigate the Temporal Relationships between Commonly Occurring Internal Medicine Diseases and Intervals for the Risk of Concurrent Disease in Canine Patients

Sequential pattern mining (SPM) is a data mining technique used for identifying common association rules in multiple sequential datasets and patterns in ordered events. In this study, we aimed to identify the relationships between commonly occurring internal medicine diseases in canine patients. We obtained medical records of dogs referred to the Konkuk University Veterinary Medicine Teaching Hospital. The data used for SPM included comorbidities and intervals between the diagnoses of internal medicine diseases. Additionally, we estimated the 3-year risk of developing an additional disease after the initial diagnosis of a commonly occurring veterinary internal medicine disease using logistic regression. We identified 547 canine patients diagnosed with ≥ 1 internal medicine disease. The SPM-based analysis assessed comorbidities and intervals for each of the five most common internal medical diseases, including hyperadrenocorticism, myxomatous mitral valve disease, canine atopic dermatitis, chronic kidney disease, and chronic pancreatitis. The highest values of the association rule were 3.01%, 6.02%, 3.9%, 4.1%, and 4.84%, and the shortest intervals were 1.64, 13.14, 5.37, 17.02, and 1.7 days, respectively. This study proposes that SPM is an effective technique for identifying common associations and temporal relationships between internal medicine diseases, and can be used to assess the probability of additional admission due to the development of the subsequent disease that may be diagnosed in canine patients. The results of this study will help veterinarians suggest appropriate preventive measures or other medical treatments for canine patients with medical conditions that have not yet been diagnosed, but are likely to develop in the short term.

Application of Radiomics in Prognosing Lung Cancer Treated with Epidermal Growth Factor Receptor Tyrosine Kinase Inhibitors: A Systematic Review and Meta-Analysis

In the context of non-small cell lung cancer (NSCLC) patients treated with EGFR tyrosine kinase inhibitors (TKIs), this research evaluated the prognostic value of CT-based radiomics. A comprehensive systematic review and meta-analysis of studies up to April 2023, which included 3111 patients, was conducted. We utilized the Quality in Prognosis Studies (QUIPS) tool and radiomics quality scoring (RQS) system to assess the quality of the included studies. Our analysis revealed a pooled hazard ratio for progression-free survival of 2.80 (95% confidence interval: 1.87-4.19), suggesting that patients with certain radiomics features had a significantly higher risk of disease progression. Additionally, we calculated the pooled Harrell's concordance index and area under the curve (AUC) values of 0.71 and 0.73, respectively, indicating good predictive performance of radiomics. Despite these promising results, further studies with consistent and robust protocols are needed to confirm the prognostic role of radiomics in NSCLC.

Prognostic value of consolidation-to-tumor ratio on computed tomography in NSCLC: a meta-analysis

BACKGROUND

Although several studies have confirmed the prognostic value of the consolidation to tumor ratio (CTR) in non-small cell lung cancer (NSCLC), there still remains controversial about it.

METHODS

We systematically searched the PubMed, Embase, and Web of Science databases from inception to April, 2022 for eligible studies that reported the correlation between CTR and prognosis in NSCLC. Hazard ratios (HRs) with 95% confidence intervals (95% CIs) were extracted and pooled to assess the overall effects. Heterogeneity was estimated by I² statistics. Subgroup analysis based on the cut-off value of CTR, country, source of HR and histology type was conducted to detect the sources of heterogeneity. Statistical analyses were performed using STATA version 12.0.

RESULTS

A total of 29 studies published between 2001 and 2022 with 10,347 patients were enrolled. The pooled results demonstrated that elevated CTR was associated with poorer overall survival (HR = 1.88, 95% CI 1.42-2.50, P < 0.01) and disease-free survival (DFS)/recurrence-free survival (RFS)/progression-free survival (PFS) (HR = 1.42, 95% CI 1.27-1.59, P < 0.01) in NSCLC. According to subgroup analysis by the cut-off value of CTR and histology type, both lung adenocarcinoma and NSCLC patients who had a higher CTR showed worse survival. Subgroup analysis stratified by country revealed that CTR was a prognostic factor for OS and DFS/RFS/PFS in Chinese, Japanese, and Turkish patients.

CONCLUSIONS

In NSCLC patients with high CTR, the prognosis was worse than that with low CTR, indicating that CTR may be a prognostic factor.

Transfer-Learning Deep Radiomics and Hand-Crafted Radiomics for Classifying Lymph Nodes from Contrast-Enhanced Computed Tomography in Lung Cancer

OBJECTIVES

Positron emission tomography (PET) is currently considered the non-invasive reference standard for lymph node (N-)staging in lung cancer. However, not all patients can undergo this diagnostic procedure due to high costs, limited availability, and additional radiation exposure. The purpose of this study was to predict the PET result from traditional contrast-enhanced computed tomography (CT) and to test different feature extraction strategies.

METHODS

In this study, 100 lung cancer patients underwent a contrast-enhanced ¹⁸F-fluorodeoxyglucose (FDG) PET/CT scan between August 2012 and December 2019. We trained machine learning models to predict FDG uptake in the subsequent PET scan. Model inputs were composed of (i) traditional "hand-crafted" radiomics features from the segmented lymph nodes, (ii) deep features derived from a pretrained EfficientNet-CNN, and (iii) a hybrid approach combining (i) and (ii).

RESULTS

In total, 2734 lymph nodes [555 (20.3%) PET-positive] from 100 patients [49% female; mean age 65, SD: 14] with lung cancer (60% adenocarcinoma, 21% plate epithelial carcinoma, 8% small-cell lung cancer) were included in this study. The area under the receiver operating characteristic curve (AUC) ranged from 0.79 to 0.87, and the scaled Brier score (SBS) ranged from 16 to 36%. The random forest model (iii) yielded the best results [AUC 0.871 (0.865-0.878), SBS 35.8 (34.2-37.2)] and had significantly higher model performance than both approaches alone (AUC: p < 0.001, z = 8.8 and z = 22.4; SBS: p < 0.001, z = 11.4 and z = 26.6, against (i) and (ii), respectively).

CONCLUSION

Both traditional radiomics features and transfer-learning deep radiomics features provide relevant and complementary information for non-invasive N-staging in lung cancer.