Enhancement of prostate cancer diagnosis by machine learning techniques: an algorithm development and validation study.

This study was undertaken to determine whether the prostate specific antigen (PSA) density (PSAD) and PSAD of the transition zone (PSADT) are useful in the detection of prostate carcinoma in Japanese men with intermediate levels of serum PSA. Two hundred and eighty-seven Japanese men with intermediate serum PSA levels (2.1 ng/mL to 10 ng/mL) underwent measurement of prostate volume by transrectal ultrasound (TRUS) and systematic biopsy under TRUS guidance. The volume of the transition zone was also measured by TRUS in 134 patients. The PSAD and PSADT were determined for each patient, and their relationship to prostate carcinoma detection was examined. Prostate carcinoma was detected in 30 of 287 patients (10.5%). Although the serum PSA levels were similar in patients with benign and malignant prostate disease (P = 0.541), the prostate volume (P 0.0009) and PSAD (P < 0.0001) differed significantly in the two groups; in the patients with prostate carcinoma, the prostate volume was smaller, and the PSAD higher, than in the patients with benign disease. At the PSAD cutoff value of 0.18 ng/mL/cm3 or greater, the sensitivity was 70% and the specificity was 67% for the diagnosis of prostate carcinoma. The PSAD was found to be significantly better in the differentiation between benign and malignant prostate disease than the serum PSA in the receiver operating characteristic analyses (P = 0.045). However, the receiver operating characteristic curve for PSAD was not significantly different compared with that for PSA in the men with negative digital rectal examination findings. Prostate carcinoma was detected in 9.0% (12 of 134) of the patients who underwent PSADT determination. Receiver operating characteristic analyses showed that PSADT was not superior to PSA in the detection of prostate carcinoma. In Japanese men with intermediate serum PSA concentrations, PSAD offers additional information useful in the detection of prostate carcinoma, but PSADT does not. Although use of PSAD may decrease the number of unnecessary biopsies, a significant number of prostate carcinomas may be overlooked. Therefore, the authors recommend that serum PSA levels continue to be used as an indicator for biopsy in Japanese men.

Gastrointestinal bleeding (GIB) is a severe and potentially life-threatening complication in patients with acute myocardial infarction (AMI), significantly affecting prognosis during hospitalization. Early identification of high-risk patients is essential to reduce complications, improve outcomes, and guide clinical decision-making. This study aimed to develop and validate a machine learning (ML)-based model for predicting in-hospital GIB in patients with AMI, identify key risk factors, and evaluate the clinical applicability of the model for risk stratification and decision support. A multicenter retrospective cohort study was conducted, including 1910 patients with AMI from the Affiliated Hospital of Guangdong Medical University (2005-2024). Patients were divided into training (n=1575) and testing (n=335) cohorts based on admission dates. For external validation, 1746 patients with AMI were included in the publicly available MIMIC-IV (Medical Information Mart for Intensive Care IV) database. Propensity score matching was adjusted for demographics, and the Boruta algorithm identified key predictors. A total of 7 ML algorithms-logistic regression, k-nearest neighbors, support vector machine, decision tree, random forest (RF), extreme gradient boosting, and neural networks-were trained using 10-fold cross-validation. The models were evaluated for the area under the receiver operating characteristic curve, accuracy, sensitivity, specificity, recall, F1-score, and decision curve analysis. Shapley additive explanations analysis ranked variable importance. Kaplan-Meier survival analysis evaluated the impact of GIB on short-term survival. Multivariate logistic regression assessed the relationship between coronary heart disease (CHD) and in-hospital GIB after adjusting for clinical variables. The RF model outperformed other ML models, achieving an area under the receiver operating characteristic curve of 0.77 in the training cohort, 0.77 in the testing cohort, and 0.75 in the validation cohort. Key predictors included red blood cell count, hemoglobin, maximal myoglobin, hematocrit, CHD, and other variables, all of which were strongly associated with GIB risk. Decision curve analysis demonstrated the clinical use of the RF model for early risk stratification. Kaplan-Meier survival analysis showed no significant differences in 7- and 15-day survival rates between patients with AMI with and without GIB (P=.83 for 7-day survival and P=.87 for 15-day survival). Multivariate logistic regression showed that CHD was an independent risk factor for in-hospital GIB (odds ratio 2.79, 95% CI 2.09-3.74). Stratified analyses by sex, age, occupation, marital status, and other subgroups consistently showed that the association between CHD and GIB remained robust across all subgroups. The ML-based RF model provides a robust and clinically applicable tool for predicting in-hospital GIB in patients with AMI. By leveraging routinely available clinical and laboratory data, the model supports early risk stratification and personalized preventive strategies.

ObjectiveTo develop and validate a generalized prediction model that can classify epidermal growth factor receptor (EGFR) mutation status in non–small cell lung cancer patients.MethodsA total of 346 patients (296 in the training cohort and 50 in the validation cohort) from four centers were included in this retrospective study. First, 1085 features were extracted using IBEX from the computed tomography images. The features were screened using the intraclass correlation coefficient, hypothesis tests and least absolute shrinkage and selection operator. Logistic regression (LR), decision tree (DT), random forest (RF), and support vector machine (SVM) were used to build a radiomics model for classification. The models were evaluated using the following metrics: area under the curve (AUC), calibration curve (CAL), decision curve analysis (DCA), concordance index (C-index), and Brier score.ResultsSixteen features were selected, and models were built using LR, DT, RF, and SVM. In the training cohort, the AUCs was .723, .842, .995, and .883; In the validation cohort, the AUCs were .658, 0567, .88, and .765. RF model with the best AUC, its CAL, C-index (training cohort=.998; validation cohort=.883), and Brier score (training cohort=.007; validation cohort=0.137) showed a satisfactory predictive accuracy; DCA indicated that the RF model has better clinical application value.ConclusionMachine learning models based on computed tomography images can be used to evaluate EGFR status in patients with non–small cell lung cancer, and the RF model outperformed LR, DT, and SVM.

Background Machine learning (ML), a subset of artificial intelligence (AI), uses algorithms to analyze data and predict outcomes without extensive human intervention. In healthcare, ML is gaining attention for enhancing patient outcomes. This study focuses on predicting additional hospital days (AHD) for patients with cervical spondylosis (CS), a condition affecting the cervical spine. The research aims to develop an ML-based nomogram model analyzing clinical and demographic factors to estimate hospital length of stay (LOS). Accurate AHD predictions enable efficient resource allocation, improved patient care, and potential cost reduction in healthcare. Methods The study selected CS patients undergoing cervical spine surgery and investigated their medical data. A total of 945 patients were recruited, with 570 males and 375 females. The mean number of LOS calculated for the total sample was 8.64 ± 3.7 days. A LOS equal to or <8.64 days was categorized as the AHD-negative group (n = 539), and a LOS > 8.64 days comprised the AHD-positive group (n = 406). The collected data was randomly divided into training and validation cohorts using a 7:3 ratio. The parameters included their general conditions, chronic diseases, preoperative clinical scores, and preoperative radiographic data including ossification of the anterior longitudinal ligament (OALL), ossification of the posterior longitudinal ligament (OPLL), cervical instability and magnetic resonance imaging T2-weighted imaging high signal (MRI T2WIHS), operative indicators and complications. ML-based models like Lasso regression, random forest (RF), and support vector machine (SVM) recursive feature elimination (SVM-RFE) were developed for predicting AHD-related risk factors. The intersections of the variables screened by the aforementioned algorithms were utilized to construct a nomogram model for predicting AHD in patients. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve and C-index were used to evaluate the performance of the nomogram. Calibration curve and decision curve analysis (DCA) were performed to test the calibration performance and clinical utility. Results For these participants, 25 statistically significant parameters were identified as risk factors for AHD. Among these, nine factors were obtained as the intersection factors of these three ML algorithms and were used to develop a nomogram model. These factors were gender, age, body mass index (BMI), American Spinal Injury Association (ASIA) scores, magnetic resonance imaging T2-weighted imaging high signal (MRI T2WIHS), operated segment, intraoperative bleeding volume, the volume of drainage, and diabetes. After model validation, the AUC was 0.753 in the training cohort and 0.777 in the validation cohort. The calibration curve exhibited a satisfactory agreement between the nomogram predictions and actual probabilities. The C-index was 0.788 (95% confidence interval: 0.73214–0.84386). On the decision curve analysis (DCA), the threshold probability of the nomogram ranged from 1 to 99% (training cohort) and 1 to 75% (validation cohort). Conclusion We successfully developed an ML model for predicting AHD in patients undergoing cervical spine surgery, showcasing its potential to support clinicians in AHD identification and enhance perioperative treatment strategies.

Comparison of Prostate Specific Antigen Concentration Versus Prostate Specific Antigen Density in the Early Detection of Prostate Cancer: Receiver Operating Characteristic Curves

Background Androgen receptor signaling inhibitors (ARSIs), including abiraterone acetate and enzalutamide, were approved for metastatic castration-resistant prostate cancer (mCRPC) based on landmark trials showing improved survival in both the pre- and post-chemotherapy settings. However, multiple real-world studies have demonstrated increased cardiovascular (CV) morbidity with these agents. Purpose To develop and validate a machine learning-based CV risk prediction model in prostate cancer (PCa) patients receiving ARSIs. Methods A nationwide cohort study was conducted utilizing data from the National Health Insurance Research Database containing the Taiwan Cancer Registry. The study population comprised 4,739 PCa patients who received abiraterone or enzalutamide between January 1, 2014, and February 28, 2022. The cohort was divided into a training set (70%, n=3,318) and a validation set (30%, n=1,421). The index date was defined as the initiation of either ARSI. A machine learning technique with random survival forest (RSF) model incorporating 16 variables was developed to predict major adverse cardiovascular events (MACEs), which were defined as a composite outcome encompassing heart failure (HF) hospitalization, myocardial infarction, ischemic stroke, and cardiovascular mortality. Results Over a mean follow-up period of 2.1 years, MACEs occurred in 10.9% and 11.3% of the training and validation cohorts, respectively. The RSF model utilized variable importance (VIMP) to rank the prognostic capability of 16 clinically relevant variables routinely available during ARSI treatment. These variables encompassed patient's age, ARSI type, androgen deprivation therapy (ADT) modality, eight comorbidities (not including cardiovascular disease because its components have been included), four prior cardiovascular events requiring hospitalization, and previous docetaxel use. The final RSF model identified five key predictive indicators: age &amp;lt;65 or ≥75 years, heart failure, stroke, hypertension, and myocardial infarction. The model exhibited robust performance, achieving an area under the curve (AUC) of 85.1% in the training set and demonstrating strong external validity with an AUC of 85.5% in the validation cohort. A positive correlation was observed between the number of risk factors and the incidence of MACEs. In the training cohort, stratification by risk factor count yielded the following 4-year MACE rates: 6.9% for patients with 0-1 risk factors, 15.3% for those with 2 risk factors, and 24.8% for individuals with ≥3 risk factors (P trend &amp;lt;0.001). This finding was consistent in the validation cohort. Conclusions This machine learning approach identified five predictors of MACEs in PCa patients receiving ARSIs. The risk stratification model provides clinicians with an evidence-based tool for individualized cardiovascular risk assessment prior to initiating ARSI therapy.Variable Importance Distribution Risk Stratification Based on Predictors

Prostate specific antigen (PSA) does not appear to have the specificity to distinguish between benign prostate hyperplasia and cancer when the PSA is low. PSA density is thought by many to improve the specificity for cancer; however, this theory remains controversial. The authors retrospectively reviewed 220 carcinomas in radical prostatectomy specimens and examined the relationship of PSA and PSA density to prostate volume, Gleason sum, and pathologic stage. Prostate specific antigen and PSA density parallel each other and do not appear to correlate statistically with displaced volume of the prostate, Gleason sum, or pathologic stage. However, PSA density in the PSA 4.1-10 ng/ml group may have conferred unique information secondary to increased variation in prostate volume. Furthermore, PSA density was associated more than PSA with carcinoma in the PSA < or = 4.0 ng/ml group. A PSA density cutoff of greater than 0.05 ng/ml/ml was accurate in the diagnosis of 94.9% of the patients with cancer. Finally, carcinomas with a Gleason sum of greater than 6 in patients with a PSA density of greater than 0.3 ng/ml/ml had a high probability of being extra-capsular at the time of surgery. Although PSA and PSA density appear to mirror each other in many ways, PSA density confers unique information and may be used as an adjunct to PSA and digital rectal examination in the detection and staging of prostate cancer. If prostate needle biopsy is performed, PSA density and Gleason sum may help identify patients who are at high risk for surgical failure.

BackgroundIn critically ill patients with cirrhosis, gastrointestinal bleeding (GIB) is a common complication that significantly impacts clinical outcomes during ICU hospitalization. Early identification of high-risk patients is crucial for preventing complications and guiding appropriate clinical interventions, which can improve treatment outcomes.ObjectiveTo develop and externally validate a machine learning model for predicting in-hospital GIB in ICU patients with cirrhosis, identify key predictors, and assess its clinical utility for risk stratification and decision-making.MethodsA retrospective cohort study was conducted, including 3,160 ICU patients diagnosed with cirrhosis from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. Patients were divided chronologically into training (n = 2,528) and testing (n = 632) cohorts based on their ICU admission dates. External validation was performed on a separate cohort of 523 ICU patients with cirrhosis extracted from the publicly available Electronic Intensive Care Unit (EICU) database. Key predictive variables were identified through a combination of the Boruta algorithm, correlation analysis, and variance inflation factor (VIF) assessment, ensuring both predictive relevance and control of multicollinearity. Six ML algorithms—logistic regression, k-nearest neighbors, support vector machine, random forest (RF), multilayer perceptron, extreme gradient boosting, and gradient boosting machine—were trained and evaluated through 10-fold cross-validation. Model performance was rigorously assessed based on the area under the receiver operating characteristic curve (AUC-ROC), accuracy, sensitivity, specificity, F1-score, calibration curves, and decision curve analysis (DCA). Shapley additive explanations (SHAP) analysis was employed to interpret and rank variable importance. Additionally, multivariable logistic regression models were constructed to elucidate the relationship between anticoagulant therapy and the incidence of in-ICU GIB after comprehensive adjustment for relevant clinical factors.ResultsAmong the ML algorithms evaluated, the RF model achieved AUC of 0.86 (95% CI: 0.84–0.88) in the training cohort and 0.72 (95% CI: 0.68–0.76) in the test cohort, with sensitivity 0.68, specificity 0.71, and precision 0.47. The key predictors identified by the model included red blood cell count, hemoglobin level, platelet count, and anticoagulant therapy, all of which were significantly associated with the risk of gastrointestinal bleeding. Decision curve analysis indicated that the RF model provides meaningful clinical utility for early risk stratification. Multivariable logistic regression further revealed that anticoagulant use independently correlated with a lower risk of in-ICU GIB (or: 0.29; 95% confidence interval: 0.24–0.34). Stratified analyses based on gender, age, weight, and additional subgroups consistently confirmed the robustness of the protective association between anticoagulant therapy and reduced GIB risk.ConclusionThe RF model demonstrated stable discrimination for predicting GIB risk in ICU patients with cirrhosis across multiple cohorts. Built on readily available clinical data, it enables timely risk stratification and informs individualized preventive interventions in critical care settings.

BACKGROUNDFor patients with portal hypertension (PH), portal vein thrombosis (PVT) is a fatal complication after splenectomy. Postoperative platelet elevation is considered the foremost reason for PVT. However, the value of postoperative platelet elevation rate (PPER) in predicting PVT has never been studied.AIMTo investigate the predictive value of PPER for PVT and establish PPER-based prediction models to early identify individuals at high risk of PVT after splenectomy.METHODSWe retrospectively reviewed 483 patients with PH related to hepatitis B virus who underwent splenectomy between July 2011 and September 2018, and they were randomized into either a training (n = 338) or a validation (n = 145) cohort. The generalized linear (GL) method, least absolute shrinkage and selection operator (LASSO), and random forest (RF) were used to construct models. The receiver operating characteristic curves (ROC), calibration curve, decision curve analysis (DCA), and clinical impact curve (CIC) were used to evaluate the robustness and clinical practicability of the GL model (GLM), LASSO model (LSM), and RF model (RFM).RESULTSMultivariate analysis exhibited that the first and third days for PPER (PPER1, PPER3) were strongly associated with PVT [odds ratio (OR): 1.78, 95% confidence interval (CI): 1.24-2.62, P = 0.002; OR: 1.43, 95%CI: 1.16-1.77, P < 0.001, respectively]. The areas under the ROC curves of the GLM, LSM, and RFM in the training cohort were 0.83 (95%CI: 0.79-0.88), 0.84 (95%CI: 0.79-0.88), and 0.84 (95%CI: 0.79-0.88), respectively; and were 0.77 (95%CI: 0.69-0.85), 0.83 (95%CI: 0.76-0.90), and 0.78 (95%CI: 0.70-0.85) in the validation cohort, respectively. The calibration curves showed satisfactory agreement between prediction by models and actual observation. DCA and CIC indicated that all models conferred high clinical net benefits.CONCLUSIONPPER1 and PPER3 are effective indicators for postoperative prediction of PVT. We have successfully developed PPER-based practical models to accurately predict PVT, which would conveniently help clinicians rapidly differentiate individuals at high risk of PVT, and thus guide the adoption of timely interventions.

Machine Learning to Predict Fascial Dehiscence after Exploratory Laparotomy Surgery

Objective Developing a follow-up strategy to reduce unnecessary prostate biopsies in patients with prostate specific antigen (PSA) ranging from 4 to 10 ng/ml based on the new model. Methods In the first stage, 436 patients with PSA ranging from 4 to 10 ng/ml who had undergone transrectal ultrasound (TRUS)-guided prostate biopsy were evaluated. PSA, free PSA (fPSA), digital rectal examination (DRE) findings, ultrasonic hypoechoic mass, ultrasonic microcalcifications, prostate volume (PV), PSA density (PSAD) and age, were regarded as predictive factors. In order to select independent predictors, a multiple logistic regression analysis involving a backward elimination selection procedure was applied. Based on the comprehensive analysis of the results, a new model to evaluate the risk of prostate cancer and an effective follow-up strategy were developed. Results Our model included PSA, PV, fPSA, age, rate of abnormal DRE findings and rate of hypoechoic masses detected by TRUS. When compared with using PSA alone, an evidently large area under the receiver-operating characteristic curve was obtained in our model(0.782 vs. 0.566). According to the degree of prostate cancer risk (PCaR), patients were divided into two groups. In the second stage of our study, prostate biopsies were implemented immediately for patients with PCaR>0.52 while the rest of the patients underwent closely follow-up. Compared with the first stage, the detectable rate of PCa in the second stage was significantly increased (33.0% vs. 21.1%, P=0.012). There was no significant difference between the two stages in the distribution of Gleason score (P=0.808). Conclusions A follow-up strategy were developed based on the new model to improve the PCa detection rate, which reduced unnecessary prostate biopsies without delaying the diagnosis and treatment of the patients. Key words: Prostate specific antigen (PSA); Model; Prostate biopsy; Prostate cancer risk (PCaR)

Most Prostate Imaging-Reporting and Data System (PI-RADS) 3 lesions do not contain clinically significant prostate cancer (CSPCa; grade group ≥2). This study was aimed at identifying clinical and magnetic resonance imaging (MRI)-derived risk fac- tors that predict CSPCa in men with PI-RADS 3 lesions. This study analyzed the detection of CSPCa in men who underwent MRI-targeted biopsy for PI-RADS 3 lesions. Multivariable logistic regression models with goodness-of-fit testing were used to identify variables associated with CSPCa. Receiver operating curves and decision curve analyses were used to estimate the clinical utility of a predictive model. Of the 1784 men reviewed, 1537 were included in the training cohort, and 247 were included in the validation cohort. The 309 men with CSPCa (17.3%) were older, had a higher prostate-specific antigen (PSA) density, and had a greater likelihood of an anteriorly located lesion than men without CSPCa (p < .01). Multivariable analysis revealed that PSA density (odds ratio [OR], 1.36; 95% confidence interval [CI], 1.05-1.85; p < .01), age (OR, 1.05; 95% CI, 1.02-1.07; p < .01), and a biopsy-naive status (OR, 1.83; 95% CI, 1.38-2.44) were independently associated with CSPCa. A prior negative biopsy was negatively associated (OR, 0.35; 95% CI, 0.24-0.50; p < .01). The application of the model to the validation cohort resulted in an area under the curve of 0.78. A predicted risk threshold of 12% could have prevented 25% of biopsies while detecting almost 95% of CSPCas with a sensitivity of 94% and a specificity of 34%. For PI-RADS 3 lesions, an elevated PSA density, older age, and a biopsy-naive status were associated with CSPCa, whereas a prior negative biopsy was negatively associated. A predictive model could prevent PI-RADS 3 biopsies while missing few CSPCas. Among men with an equivocal lesion (Prostate Imaging-Reporting and Data System 3) on multiparametric magnetic resonance imaging (mpMRI), those who are older, those who have a higher prostate-specific antigen density, and those who have never had a biopsy before are at higher risk for having clinically significant prostate cancer (CSPCa) on subsequent biopsy. However, men with at least one negative biopsy have a lower risk of CSPCa. A new predictive model can greatly reduce the need to biopsy equivocal lesions noted on mpMRI while missing only a few cases of CSPCa.

Detection of prostate cancer--experience of seven years in KMUH and review of literature.

Objective: To develop predictive models for contrast induced acute kidney injury (CI-AKI) among acute myocardial infarction (AMI) patients treated invasively.Methods: Patients with AMI who underwent angiography therapy were enrolled and randomly divided into training cohort (75%) and validation cohort (25%). Machine learning algorithms were used to construct predictive models for CI-AKI. The predictive models were tested in a validation cohort.Results: A total of 1,495 patients with AMI were included. Of all the patients, 226 (15.1%) cases developed CI-AKI. In the validation cohort, Random Forest (RF) model with top 15 variables reached an area under the curve (AUC) of 0.82 (95% CI: 0.76–0.87), while the best logistic model had an AUC of 0.69 (95% CI: 0.62–0.76). ACEF (age, creatinine, and ejection fraction) model reached an AUC of 0.62 (95% CI: 0.53–0.71). RF model with top 15 variables achieved a high recall rate of 71.9% and an accuracy of 73.5% in the validation group. Random Forest model significantly outperformed logistic regression in every comparison.Conclusions: Machine learning algorithms especially Random Forest algorithm improves the accuracy of risk stratifying patients with AMI and should be used to accurately identify the risk of CI-AKI in AMI patients.

This study was undertaken to assess the utility of prostate specific antigen (PSA) and PSA density (PSAD) in discriminating between benign and malignant prostate disease in the Kuwaiti Arab population. A total of 100 consecutive patients suspected of having prostate cancer because of serum PSA > 4 ng/ml, or detection of a prostatic nodule on rectal examination were further investigated by determination of PSAD, TRUS of prostate, sexant prostatic biopsy and histological analysis to establish the correct diagnosis. Other diagnostic measures included the determination of the area under the receiver operating characteristic (ROC) curve, sensitivity and specificity. Of the 100 prostate biopsies that were performed, 33 cases were confirmed to be prostate cancer and 67 were described as benign lesions comprising benign prostatic hyperplasia (BPH) with or without prostatitis. The age range for patients with prostate cancer was 42-90 years, and 52-90 years for those without prostate cancer. The mean prostate volume was 58.82 cc (range 9-177 cc) and 62.60 cc (range 15-140 cc), the mean PSA value was 36.65 ng/ml (range 5.8-200 ng/ml) and 16.49 ng/ml (range 1.4-46.0 ng/ml), while the mean PSAD was 0.92 (range 0.046-5.714) and 0.452 (range 0.034-2.294) for patients with prostate cancer and patients without prostate cancer respectively. Patients with PSA less than 4 ng/ml (3 cases) all had benign prostate lesions, and 7 cases with PSA more than 50 ng/ml all had prostate cancer and were excluded because values above 50 ng/ml have close to 100% specificity for prostate cancer. Further analysis was done on the remaining 90 cases which were patients with a PSA between 4 and 50 ng/ml. The discriminating power of serum PSA for detecting prostate cancer as estimated by the area under ROC was 0.686 while that for PSAD was 0.732. The maximum likelihood for a positive PSA was at a PSAD cut-off point of 0.32. For the PSA cut-off point of l0 ng/ml, the sensitivity was 80%, and specificity was 42.2%. For the PSAD cut-off point of 0.32, the sensitivity was 58% and the specificity 76.6%. Determination of PSAD is not a useful adjunct to serum PSA values in the range of 10-50 ng/ ml in our population. PSAD value less than 0.32 with PSA less than l0 ng/ml strongly suggests benign disease.