Objective:By analyzing peripheral blood indicators, we constructed and validated a novel lung cancer prediction model using machine learning algorithms for riskassessment of lung cancer.Methods:A retrospective case-control design was conducted on the clinical data of 194 newly diagnosed lung cancer patients [mean age: (66.80±9.09) years, 126 males and 68 females] admitted to Qilu Second Hospital of Shandong University between January 9, 2020, and December 31, 2024, serving as the case group. During the same period, 290 healthy individuals undergoing physical examinations [mean age: (61.18±14.31) years, 155 males and 135 females J were enrolled as the control group. A total of 46 peripheral blood indicators-including routine blood tests, coagulation parameters, liver function markers, and tumor-related indices-along with two basic characteristics (age and sex) were included in the analysis. Eleven machinelearning algorithms including logistic regression, randomforest, support vector classifier, extreme gradient boosting, gradient boosting decision tree, decision tree, multilayer perceptron, linear discriminant analysis, adaptive boosting, Gaussian naive Bayes and light gradient-boosting machine-were trained for early diagnosis of lung-cancer.Model performance was evaluated by the area under the ROC, accuracy, positive predictive value, negative predictive value, F1-score and 95% confidence interval (95% CI). The best performing algorithm was selected, and feature importance was ranked with Shapley Additive Planation(SHAP) values. Results:The support-vector classifier achieved the best performance for predicting lung-cancer risk (AUC=0.974; 95 % CI 0.951-0.989) and was retained for final model establishment. After 20 rounds of stratified 10-fold cross-validation the mean AUC was 0.950; learning-curve, decision-curve and calibration analyses confirmed its superior generalizability, clinical utility and calibration.SHAPley additive explanations and decision-tree feature importance consistently identified neuron-specific enolase, carcinoembryonic antigen, and squamous-cell carcinoma antigen as the three most critical predictors of lung-cancer risk. Conclusion:An SVM-based lung cancer prediction model was successfully established to determine the risk of developing lung cancer.

Objective:To identify the relevant factors for the first-course remission of acute myeloid leukemia (AML) and to develop a predictive model as well as assess its predictive capability.Methods:Clinical data of 749 patients newly diagnosed with AML admitted to the Department of Hematology, the First Affiliated Hospital, Zhejiang University, School of Medicine from January 1, 2019, to April 30, 2023, were collected and randomly divided into training and validation sets. Multivariate logistic regression analysis was conducted to determine variables associated with complete remission in the first course of induction therapy, and a predictive model was established based on these variables. The receiver operating characteristic (ROC) curve of the predictive model was plotted, and the area under the curve (AUC) was calculated.Results:The indicators predicting the first remission course included peripheral blood white blood cell count during onset, CBF::MYH11 fusion gene, CEBPA bZIP region mutation, myelodysplastic syndrome-related gene mutation, and induction chemotherapy regimen selection as independent factors for the first remission course. The model’s area under the training and validation curves was 0.738 (95% CI: 0.696-0.780) and 0.726 (95% CI: 0.650-0.801), respectively. The Hosmer-Lemeshow test results yielded P-values of 0.993 and 0.335, respectively. Conclusion:In this study, the developed model demonstrates a strong predictive capability for the efficacy of the first course of patients with AML, providing valuable guidance to clinicians in assessing patient prognosis and selecting appropriate treatment strategies.

Objective: The CT/MRI Liver Imaging Reporting and Data System (LI-RADS) demonstrates high specificity with relatively limited sensitivity for diagnosing hepatocellular carcinoma (HCC) in high-risk patients. This study aimed to explore the possibility of improving sensitivity by combining CT/MRI LI-RADS v2018 with second-line contrast-enhanced ultrasound (CEUS) LI-RADS v2017 using sulfur hexafluoride (SHF) or perfluorobutane (PFB). Materials and Methods: This retrospective analysis of prospectively collected multicenter data included high-risk patients with treatment-naive hepatic observations. The reference standard was pathological confirmation or a composite reference standard (only for benign lesions). Each participant underwent concurrent CT/MRI, SHF-enhanced US, and PFB-enhanced US examinations. The diagnostic performances for HCC of CT/MRI LI-RADS alone and three combination strategies (combining CT/ MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or a modified algorithm incorporating the Kupffer-phase findings for PFB [modified PFB]) were evaluated. For the three combination strategies, apart from the CT/MRI LR-5 criteria, HCC was diagnosed if CT/MRI LR-3 or LR-4 observations met the LR-5 criteria using LI-RADS SHF, LI-RADS PFB, or modified PFB. Results: In total, 281 participants (237 males; mean age, 55 ± 11 years) with 306 observations (227 HCCs, 40 non-HCC malignancies, and 39 benign lesions) were included. Using LI-RADS SHF, LI-RADS PFB, and modified PFB, 20, 23, and 31 CT/MRI LR-3/4 observations, respectively, were reclassified as LR-5, and all were pathologically confirmed as HCCs. Compared to CT/MRI LI-RADS alone (74%, 95% confidence interval [CI]: 68%–79%), the three combination strategies combining CT/MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or modified PFB increased sensitivity (83% [95% CI: 77%–87%], 84% [95% CI: 79%–89%], 88% [95% CI: 83%–92%], respectively; all P < 0.001), while maintaining the specificity at 92% (95% CI: 84%–97%). Conclusion The combination of CT/MRI LI-RADS with second-line CEUS using SHF or PFB improved the sensitivity of HCC diagnosis without compromising specificity.

Objective: The CT/MRI Liver Imaging Reporting and Data System (LI-RADS) demonstrates high specificity with relatively limited sensitivity for diagnosing hepatocellular carcinoma (HCC) in high-risk patients. This study aimed to explore the possibility of improving sensitivity by combining CT/MRI LI-RADS v2018 with second-line contrast-enhanced ultrasound (CEUS) LI-RADS v2017 using sulfur hexafluoride (SHF) or perfluorobutane (PFB). Materials and Methods: This retrospective analysis of prospectively collected multicenter data included high-risk patients with treatment-naive hepatic observations. The reference standard was pathological confirmation or a composite reference standard (only for benign lesions). Each participant underwent concurrent CT/MRI, SHF-enhanced US, and PFB-enhanced US examinations. The diagnostic performances for HCC of CT/MRI LI-RADS alone and three combination strategies (combining CT/ MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or a modified algorithm incorporating the Kupffer-phase findings for PFB [modified PFB]) were evaluated. For the three combination strategies, apart from the CT/MRI LR-5 criteria, HCC was diagnosed if CT/MRI LR-3 or LR-4 observations met the LR-5 criteria using LI-RADS SHF, LI-RADS PFB, or modified PFB. Results: In total, 281 participants (237 males; mean age, 55 ± 11 years) with 306 observations (227 HCCs, 40 non-HCC malignancies, and 39 benign lesions) were included. Using LI-RADS SHF, LI-RADS PFB, and modified PFB, 20, 23, and 31 CT/MRI LR-3/4 observations, respectively, were reclassified as LR-5, and all were pathologically confirmed as HCCs. Compared to CT/MRI LI-RADS alone (74%, 95% confidence interval [CI]: 68%–79%), the three combination strategies combining CT/MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or modified PFB increased sensitivity (83% [95% CI: 77%–87%], 84% [95% CI: 79%–89%], 88% [95% CI: 83%–92%], respectively; all P < 0.001), while maintaining the specificity at 92% (95% CI: 84%–97%). Conclusion The combination of CT/MRI LI-RADS with second-line CEUS using SHF or PFB improved the sensitivity of HCC diagnosis without compromising specificity.

Objective: The CT/MRI Liver Imaging Reporting and Data System (LI-RADS) demonstrates high specificity with relatively limited sensitivity for diagnosing hepatocellular carcinoma (HCC) in high-risk patients. This study aimed to explore the possibility of improving sensitivity by combining CT/MRI LI-RADS v2018 with second-line contrast-enhanced ultrasound (CEUS) LI-RADS v2017 using sulfur hexafluoride (SHF) or perfluorobutane (PFB). Materials and Methods: This retrospective analysis of prospectively collected multicenter data included high-risk patients with treatment-naive hepatic observations. The reference standard was pathological confirmation or a composite reference standard (only for benign lesions). Each participant underwent concurrent CT/MRI, SHF-enhanced US, and PFB-enhanced US examinations. The diagnostic performances for HCC of CT/MRI LI-RADS alone and three combination strategies (combining CT/ MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or a modified algorithm incorporating the Kupffer-phase findings for PFB [modified PFB]) were evaluated. For the three combination strategies, apart from the CT/MRI LR-5 criteria, HCC was diagnosed if CT/MRI LR-3 or LR-4 observations met the LR-5 criteria using LI-RADS SHF, LI-RADS PFB, or modified PFB. Results: In total, 281 participants (237 males; mean age, 55 ± 11 years) with 306 observations (227 HCCs, 40 non-HCC malignancies, and 39 benign lesions) were included. Using LI-RADS SHF, LI-RADS PFB, and modified PFB, 20, 23, and 31 CT/MRI LR-3/4 observations, respectively, were reclassified as LR-5, and all were pathologically confirmed as HCCs. Compared to CT/MRI LI-RADS alone (74%, 95% confidence interval [CI]: 68%–79%), the three combination strategies combining CT/MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or modified PFB increased sensitivity (83% [95% CI: 77%–87%], 84% [95% CI: 79%–89%], 88% [95% CI: 83%–92%], respectively; all P < 0.001), while maintaining the specificity at 92% (95% CI: 84%–97%). Conclusion The combination of CT/MRI LI-RADS with second-line CEUS using SHF or PFB improved the sensitivity of HCC diagnosis without compromising specificity.

Objective: The CT/MRI Liver Imaging Reporting and Data System (LI-RADS) demonstrates high specificity with relatively limited sensitivity for diagnosing hepatocellular carcinoma (HCC) in high-risk patients. This study aimed to explore the possibility of improving sensitivity by combining CT/MRI LI-RADS v2018 with second-line contrast-enhanced ultrasound (CEUS) LI-RADS v2017 using sulfur hexafluoride (SHF) or perfluorobutane (PFB). Materials and Methods: This retrospective analysis of prospectively collected multicenter data included high-risk patients with treatment-naive hepatic observations. The reference standard was pathological confirmation or a composite reference standard (only for benign lesions). Each participant underwent concurrent CT/MRI, SHF-enhanced US, and PFB-enhanced US examinations. The diagnostic performances for HCC of CT/MRI LI-RADS alone and three combination strategies (combining CT/ MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or a modified algorithm incorporating the Kupffer-phase findings for PFB [modified PFB]) were evaluated. For the three combination strategies, apart from the CT/MRI LR-5 criteria, HCC was diagnosed if CT/MRI LR-3 or LR-4 observations met the LR-5 criteria using LI-RADS SHF, LI-RADS PFB, or modified PFB. Results: In total, 281 participants (237 males; mean age, 55 ± 11 years) with 306 observations (227 HCCs, 40 non-HCC malignancies, and 39 benign lesions) were included. Using LI-RADS SHF, LI-RADS PFB, and modified PFB, 20, 23, and 31 CT/MRI LR-3/4 observations, respectively, were reclassified as LR-5, and all were pathologically confirmed as HCCs. Compared to CT/MRI LI-RADS alone (74%, 95% confidence interval [CI]: 68%–79%), the three combination strategies combining CT/MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or modified PFB increased sensitivity (83% [95% CI: 77%–87%], 84% [95% CI: 79%–89%], 88% [95% CI: 83%–92%], respectively; all P < 0.001), while maintaining the specificity at 92% (95% CI: 84%–97%). Conclusion The combination of CT/MRI LI-RADS with second-line CEUS using SHF or PFB improved the sensitivity of HCC diagnosis without compromising specificity.

Objective: The CT/MRI Liver Imaging Reporting and Data System (LI-RADS) demonstrates high specificity with relatively limited sensitivity for diagnosing hepatocellular carcinoma (HCC) in high-risk patients. This study aimed to explore the possibility of improving sensitivity by combining CT/MRI LI-RADS v2018 with second-line contrast-enhanced ultrasound (CEUS) LI-RADS v2017 using sulfur hexafluoride (SHF) or perfluorobutane (PFB). Materials and Methods: This retrospective analysis of prospectively collected multicenter data included high-risk patients with treatment-naive hepatic observations. The reference standard was pathological confirmation or a composite reference standard (only for benign lesions). Each participant underwent concurrent CT/MRI, SHF-enhanced US, and PFB-enhanced US examinations. The diagnostic performances for HCC of CT/MRI LI-RADS alone and three combination strategies (combining CT/ MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or a modified algorithm incorporating the Kupffer-phase findings for PFB [modified PFB]) were evaluated. For the three combination strategies, apart from the CT/MRI LR-5 criteria, HCC was diagnosed if CT/MRI LR-3 or LR-4 observations met the LR-5 criteria using LI-RADS SHF, LI-RADS PFB, or modified PFB. Results: In total, 281 participants (237 males; mean age, 55 ± 11 years) with 306 observations (227 HCCs, 40 non-HCC malignancies, and 39 benign lesions) were included. Using LI-RADS SHF, LI-RADS PFB, and modified PFB, 20, 23, and 31 CT/MRI LR-3/4 observations, respectively, were reclassified as LR-5, and all were pathologically confirmed as HCCs. Compared to CT/MRI LI-RADS alone (74%, 95% confidence interval [CI]: 68%–79%), the three combination strategies combining CT/MRI LI-RADS with either LI-RADS SHF, LI-RADS PFB, or modified PFB increased sensitivity (83% [95% CI: 77%–87%], 84% [95% CI: 79%–89%], 88% [95% CI: 83%–92%], respectively; all P < 0.001), while maintaining the specificity at 92% (95% CI: 84%–97%). Conclusion The combination of CT/MRI LI-RADS with second-line CEUS using SHF or PFB improved the sensitivity of HCC diagnosis without compromising specificity.

Objective:To identify the relevant factors for the first-course remission of acute myeloid leukemia (AML) and to develop a predictive model as well as assess its predictive capability.Methods:Clinical data of 749 patients newly diagnosed with AML admitted to the Department of Hematology, the First Affiliated Hospital, Zhejiang University, School of Medicine from January 1, 2019, to April 30, 2023, were collected and randomly divided into training and validation sets. Multivariate logistic regression analysis was conducted to determine variables associated with complete remission in the first course of induction therapy, and a predictive model was established based on these variables. The receiver operating characteristic (ROC) curve of the predictive model was plotted, and the area under the curve (AUC) was calculated.Results:The indicators predicting the first remission course included peripheral blood white blood cell count during onset, CBF::MYH11 fusion gene, CEBPA bZIP region mutation, myelodysplastic syndrome-related gene mutation, and induction chemotherapy regimen selection as independent factors for the first remission course. The model’s area under the training and validation curves was 0.738 (95% CI: 0.696-0.780) and 0.726 (95% CI: 0.650-0.801), respectively. The Hosmer-Lemeshow test results yielded P-values of 0.993 and 0.335, respectively. Conclusion:In this study, the developed model demonstrates a strong predictive capability for the efficacy of the first course of patients with AML, providing valuable guidance to clinicians in assessing patient prognosis and selecting appropriate treatment strategies.

Objective:By analyzing peripheral blood indicators, we constructed and validated a novel lung cancer prediction model using machine learning algorithms for riskassessment of lung cancer.Methods:A retrospective case-control design was conducted on the clinical data of 194 newly diagnosed lung cancer patients [mean age: (66.80±9.09) years, 126 males and 68 females] admitted to Qilu Second Hospital of Shandong University between January 9, 2020, and December 31, 2024, serving as the case group. During the same period, 290 healthy individuals undergoing physical examinations [mean age: (61.18±14.31) years, 155 males and 135 females J were enrolled as the control group. A total of 46 peripheral blood indicators-including routine blood tests, coagulation parameters, liver function markers, and tumor-related indices-along with two basic characteristics (age and sex) were included in the analysis. Eleven machinelearning algorithms including logistic regression, randomforest, support vector classifier, extreme gradient boosting, gradient boosting decision tree, decision tree, multilayer perceptron, linear discriminant analysis, adaptive boosting, Gaussian naive Bayes and light gradient-boosting machine-were trained for early diagnosis of lung-cancer.Model performance was evaluated by the area under the ROC, accuracy, positive predictive value, negative predictive value, F1-score and 95% confidence interval (95% CI). The best performing algorithm was selected, and feature importance was ranked with Shapley Additive Planation(SHAP) values. Results:The support-vector classifier achieved the best performance for predicting lung-cancer risk (AUC=0.974; 95 % CI 0.951-0.989) and was retained for final model establishment. After 20 rounds of stratified 10-fold cross-validation the mean AUC was 0.950; learning-curve, decision-curve and calibration analyses confirmed its superior generalizability, clinical utility and calibration.SHAPley additive explanations and decision-tree feature importance consistently identified neuron-specific enolase, carcinoembryonic antigen, and squamous-cell carcinoma antigen as the three most critical predictors of lung-cancer risk. Conclusion:An SVM-based lung cancer prediction model was successfully established to determine the risk of developing lung cancer.