Breast cancer (BC) is the most frequently diagnosed malignancy worldwide and a leading cause of death in women. Machine learning (ML) enables data-driven risk stratification and outcome prediction, potentially improving screening, treatment choices, and overall care. This scoping review cataloged ML models used to predict BC risk and post-treatment outcomes. A PubMed search using MeSH terms (“breast neoplasms,” “breast cancer,” “machine learning,” and “predictive value of tests”) retrieved 44 records, with 9 added by manual searching; 24 studies were included. We excluded 29 reports limited to imaging/radiomics or biomolecular/omics, plus other ineligible items. Model usage among included studies: 13 random forest (RF); 13 support vector machine; 7 decision tree; 5 k-nearest neighbors; 4 deep learning; 3 naïve Bayes; 2 artificial neural networks. Average RF accuracy surpassed 90% in most reports, indicating RF is both common and consistently strong for predictive modeling, supporting targeted clinical translation.
breast neoplasm ; machine learning ; predictive value of tests

INTRODUCTION: Obstructive sleep apnoea (OSA) is a serious but underdiagnosed condition. Demand for the gold standard diagnostic polysomnogram (PSG) far exceeds its availability. More efficient diagnostic methods are needed, even in tertiary settings. Machine learning (ML) models have strengths in disease prediction and early diagnosis. We explored the use of ML with oximetry, demographic and anthropometric data to diagnose OSA. METHODS: A total of 2,996 patients were included for modelling and divided into test and training sets. Seven commonly used supervised learning algorithms were trained with the data. Sensitivity (recall), specificity, positive predictive value (PPV) (precision), negative predictive value, area under the receiver operating characteristic curve (AUC) and F1 measure were reported for each model. RESULTS: In the best performing four-class model (neural network model predicting no, mild, moderate or severe OSA), a prediction of moderate and/or severe disease had a combined PPV of 94%; one out of 335 patients had no OSA and 19 had mild OSA. In the best performing two-class model (logistic regression model predicting no-mild vs. moderate-severe OSA), the PPV for moderate-severe OSA was 92%; two out of 350 patients had no OSA and 26 had mild OSA. CONCLUSION Our study showed that the prediction of moderate-severe OSA in a tertiary setting with an ML approach is a viable option to facilitate early identification of OSA. Prospective studies with home-based oximeters and analysis of other oximetry variables are the next steps towards formal implementation.
Humans ; Oximetry/methods* ; Sleep Apnea, Obstructive/diagnosis* ; Male ; Female ; Middle Aged ; Machine Learning ; Polysomnography ; Adult ; Anthropometry ; ROC Curve ; Aged ; Algorithms ; Predictive Value of Tests ; Sensitivity and Specificity ; Neural Networks, Computer ; Demography

OBJECTIVES: The diagnostic value of ultrasonographic quantitative indicators of umbilical cord coiling, such as the umbilical coiling index (UCI) and pitch value, in identifying hypercoiling and predicting adverse pregnancy outcomes remains controversial. This study aims to evaluate the predictive value of UCI, pitch value, and the cerebroplacental ratio in pregnancies complicated by umbilical cord hypercoiling. METHODS: Pregnant women with densely coiled umbilical cords identified by routine obstetric ultrasound at Changsha Maternal and Child Health Hospital between November 2022 and November 2024 were enrolled. Complete clinical data, including UCI, pitch value, and cerebroplacental ratio (CPR), were collected. Pregnancy outcome scores were calculated, and newborns were categorized into the normal outcome group (n=177) and adverse outcome group (n=85), with the latter further subdivided into mild (n=51), moderate (n=19), and severe (n=15) subgroups. Differences in baseline data, UCI, pitch value, and incidence of CRP<1 were compared between groups and among subgroups. Correlations between UCI, pitch value, and adverse pregnancy outcomes were analyzed. Receiver operating characteristic (ROC) curve were used to assess the predictive performance of UCI, pitch value, CPR<1, and their combinations. RESULTS: Compared with the normal outcome group, the adverse outcome group had higher age, parity, parity, incidence of CPR<1, and UCI, while gestational age at delivery and pitch values were lower (all P<0.05). The incidence of obesity, gestational diabetes mellitus, and hypertensive disorders of pregnancy did not differ significantly between the 2 groups (all P>0.05). The normal outcome group showed lower UCI and higher pitch values than all 3 adverse outcome subgroups (all P<0.05), while differences among the 3 adverse subgroups were not significant (all P>0.05). UCI was positively correlated with adverse pregnancy outcomes (r_s=0.350, P<0.05), whereas pitch value was negatively correlated (r_s=-0.286, P<0.05). ROC curve analysis showed that the area under the curve (AUC) values for predicting adverse outcomes were 0.837 for UCI, 0.886 for pitch value, and 0.610 for CPR<1, with sensitivities of 77.6%, 82.4%, and 27.1% and specificities of 78.5%, 83.6%, and 94.9%, respectively. The combined UCI+CPR<1 and pitch value+CPR<1 models yielded AUCs of 0.841 and 0.886, with sensitivities of 78.8% and 81.2% and specificities of 78.5% and 84.2%, respectively. No significant differences were found between the AUCs of UCI and pitch value (P>0.05), but both outperformed CPR<1 alone (both P<0.001). The combined models showed no significant improvement over UCI or pitch value alone (both P>0.05), though both were superior to CPR<1 alone (both P<0.001). CONCLUSIONS Most umbilical cord hypercoiling cases had favorable outcomes, with UCI, pitch value, CPR<1 and their combinations demonstrating significant predictive value for adverse pregnancy outcomes.
Humans ; Female ; Pregnancy ; Pregnancy Outcome ; Adult ; Ultrasonography, Prenatal/methods* ; Umbilical Cord/diagnostic imaging* ; Hemodynamics ; Predictive Value of Tests ; Infant, Newborn ; ROC Curve

PURPOSE: Ultrasonography has been used increasingly in orthopaedic practice credited to its low cost, easy accessibility, non-invasiveness, reproducibility, and safety from radiation. The purpose of this study was to test the validity and efficacy of ultrasonography as an adjunct in the assessment of fracture healing in long bones treated with intramedullary interlocking devices and its predictive value in determining the need for a secondary surgical procedure. METHODS: This was a descriptive longitudinal study of 40 skeletally mature patients from November 2016 to February 2019, who sustained long bone fractures of the tibia or femur treated using intramedullary interlocking nails. Patients with comminuted and segmental fracture patterns were excluded from the study. Each patient was evaluated at 6- and 12-week post-surgery using standard orthogonal radiographs and ultrasonography to assess fracture healing. Patients were then followed up until fracture union. Quantitative data was analyzed using frequency statistics and descriptive data with inferential statistics. RESULTS: Ultrasonography predicted 87.5% union and 12.5% delayed or non-union as early as 6 weeks after surgery, while radiographs predicted 22.5% union as late as 3 months of follow-up. The sensitivity and specificity of ultrasonography in assessing fracture healing were 100% and 97.2%, respectively, with a positive predictive value of 80.0%. Vascular resistance index was less than 0.5 in all patients who developed delayed or non-union. CONCLUSION Ultrasonography is able to predict fracture outcomes much earlier than standardized radiographs with comparable sensitivity and specificity. Vascular resistance index is an objective parameter in assessing callus quality and predicting fracture outcomes.
Humans ; Fracture Fixation, Intramedullary/methods* ; Male ; Female ; Ultrasonography/methods* ; Adult ; Fracture Healing ; Tibial Fractures/diagnostic imaging* ; Middle Aged ; Femoral Fractures/diagnostic imaging* ; Longitudinal Studies ; Bone Nails ; Young Adult ; Predictive Value of Tests ; Aged ; Treatment Outcome ; Adolescent

PURPOSE: To analyze the diagnostic efficacy of computed tomography angiography compared to digital cerebral angiography for the diagnosis of traumatic aneurysms (TAs) associated with combat-related penetrating head injuries and propose the most suitable angiography protocol in this clinical context. METHODS: A retrospective analysis was conducted on patients admitted to the neurosurgical clinic for penetrating traumatic brain injuries between February, 2022 and July, 2024, for whom both cerebral multidetector computed tomography angiography (MCTA) and digital cerebral angiography (DCA) were available. The inclusion were patients (1) with penetrating head injuries, (2) with missile trajectory traverses through the Sylvian or great longitudinal fissure, (3) basal cisterns with/or major subarachnoid hemorrhage. The sensitivity, specificity, positive predictive value, and negative predictive value of MCTA were calculated. DCA was considered as the gold standard of diagnosis. The sensitivity, specificity, positive predictive value, and negative predictive value of MCTA were calculated. Descriptive statistics and nonparametric statistics were used to analyze the study results and their differences, respectively. RESULTS: A total of 40 patients with 45 TAs were included in the study. Of these, 26 patients (65.0%) were found to have aneurysms on MCTA. The median diameter of the aneurysms diagnosed by MCTA was 4.9 (3.6, 4.8) mm (range of 2.5 - 10.4 mm). However, the mean diameter of TAs not detected by MCTA but diagnosed by DCA was (3.0 ± 1.3) mm (range of 1.3 - 4.9 mm). MCTA demonstrated sensitivity and specificity of 35.5% and 99.5%, respectively, with positive and negative predictive values of 92.3% and 90.7%. CONCLUSIONS A low sensitivity of MCTA for the diagnosis of TAs associated with combat-related penetrating head injuries was reported. When MCTA is inconclusive in the setting of radiologic predictors of cerebral artery injury, DSA may be required.
Humans ; Male ; Retrospective Studies ; Adult ; Multidetector Computed Tomography/methods* ; Intracranial Aneurysm/etiology* ; Computed Tomography Angiography/methods* ; Female ; Head Injuries, Penetrating/diagnostic imaging* ; Middle Aged ; Cerebral Angiography/methods* ; Predictive Value of Tests ; Sensitivity and Specificity ; Young Adult

PURPOSE: To screen laboratory markers with predictive value in early spinal surgical site infections (SSI) that are diagnosed within 30 days postoperatively. METHODS: Patients who underwent surgical treatment for internal spinal fixation between March 2022 and March 2023 in our hospital were retrospectively studied. The inclusion criteria were aged >18 years, undergoing internal fixation surgery, complete medical records with >30 days of postoperative follow-up, diagnosis was made within 30 days postoperatively, and an informed consent form was obtained. The exclusion criteria were abnormal white blood cell count or neutrophil percentage in the preoperative blood routine and combined diseases that may affect the C-reactive protein (CRP) or procalcitonin (PCT) values, including lower respiratory tract infection, renal insufficiency, and liver disease. We collected patients' personal information, surgical information, and blood laboratory data, including CRP, PCT, lymphocyte-neutrophil ratio, platelet-neutrophil ratio, and routine blood tests on preoperative and postoperative days 3, 5, and 7, from these patients. These data were statistically analyzed to determine which laboratory markers were statistically significant. The diagnostic value and optimal diagnostic threshold of these laboratory markers were further determined by receiver operating characteristic curve analysis. RESULTS: A total of 106 patients were enrolled in this study, of whom 8 patients were diagnosed with early SSI. A total of 4 laboratory markers were screened, namely, CRP on postoperative day 7 (optimal diagnostic threshold of ≥64.1 mg/L, sensitivity of 100%, specificity of 76.5%, area under the curve (AUC) of 0.908), PCT on postoperative day 7 (optimal diagnostic threshold of ≥0.2 ng/mL, sensitivity of 87.5%, specificity of 94.1%, AUC of 0.967), lymphocyte count on postoperative day 5 (optimal diagnostic threshold of ≤0.67 × 10⁹/L, sensitivity of 50%, specificity of 95.9%, AUC of 0.760), and lymphocyte count on postoperative day 7 (optimal diagnostic threshold of ≤1.32 × 10⁹/L, sensitivity of 87.5%, specificity of 55.1%, AUC of 0.721). CONCLUSION We concluded that CRP and PCT levels on postoperative day 7 and lymphocyte counts on postoperative days 5 and 7 are useful markers in screening for early spinal SSI.
Humans ; Retrospective Studies ; Male ; Female ; Biomarkers/blood* ; Middle Aged ; C-Reactive Protein/analysis* ; Surgical Wound Infection/blood* ; Procalcitonin/blood* ; Adult ; Aged ; Postoperative Period ; ROC Curve ; Predictive Value of Tests ; Spine/surgery*

OBJECTIVES: To compare serum insulin-like growth factor 1 (IGF-1) and peak growth hormone (GH) levels between girls with isolated premature thelarche (IPT) and central precocious puberty (CPP), to construct a prediction model for progression from IPT to CPP, and to assess its diagnostic value. METHODS: Girls diagnosed with IPT (n=111) between January 2022 and August 2023 at the China-Japan Friendship Hospital and the Xinjiang Production and Construction Corps Hospital were retrospectively included. According to follow-up outcomes, participants were categorized into a CPP group (35 cases) and an IPT group (36 cases). A clinical prediction model for progression to CPP was constructed by multivariable logistic regression, and the contributions of IGF-1 and peak GH were evaluated. Restricted cubic spline analysis was used to assess the dose-response relationships of IGF-1 and peak GH with CPP. Decision curve analysis was applied to evaluate clinical utility. RESULTS: IGF-1 and peak GH were higher in the CPP group than in the IPT group (P<0.05). Compared with model 1 (without IGF-1 and peak GH), model 2 (with IGF-1 and peak GH) showed significantly higher area under the curve, integrated discrimination improvement, and net reclassification improvement (all P<0.05). Model 2 (χ 2=6.054, P=0.889) also demonstrated better goodness-of-fit than model 1 (χ 2=7.717, P=0.634). Nonlinear dose-response relationships were observed for peak GH and IGF-1 with CPP (P for overall trend <0.05; P for nonlinearity <0.05). Decision curve analysis indicated that combined prediction using IGF-1 and peak GH provided greater net benefit than either biomarker alone. CONCLUSIONS Peak GH and IGF-1 are closely associated with progression from IPT to CPP in girls. A clinical prediction model incorporating peak GH and IGF-1 can improve prediction of progression to CPP and yield higher net benefit.
Humans ; Female ; Puberty, Precocious/etiology* ; Insulin-Like Growth Factor I/analysis* ; Child ; Retrospective Studies ; Human Growth Hormone/blood* ; Predictive Value of Tests ; Child, Preschool ; Logistic Models

OBJECTIVE: To explore the predictive value of age, D-Dimer and fibrinogen (FIB) for non-thrombotic. METHODS: A retrospective analysis was conducted on a total of 1 384 coagulation test cases from January to August 2024 at Nanning No. 8 People's Hospital. Among them, the control group comprised 400 non-thrombotic cases with D-Dimer test results within the reference range. The thrombotic group comprised 57 clinically diagnosed thrombotic patients. The research group comprised 927 non-thrombotic cases with D-Dimer levels exceeding the reference range. The diagnosis treatment records, age information, plasma D-Dimer, and FIB test results of each group were collected. The changes and correlations of age, D-Dimer, and FIB indicators were compared and analyzed among the three groups. A new combination factor was generated by fitting a Logistic binary regression model. ROC curves were used to evaluate the predictive value of each index for non-thrombotic disease in both the research group and the thrombotic group. RESULTS: Compared with the control group, the thrombotic group and the research group had significantly higher age, D-Dimer, and FIB levels (P < 0.001). Further comparative analysis showed that the research group had significantly lower age and D-Dimer levels than the thrombotic group, the FIB level was significantly higher than that of the thrombotic group (P < 0.001). Spearman correlation analysis showed that the correlation coefficient between age and D-Dimer in the research group was higher than that in the control group and thrombotic group (P < 0.01), the thrombotic group had the highest negative correlation coefficient between FIB and D-Dimer (P < 0.01). The ROC curve analysis results showed that the AUC values of age, plasma D-dimer, and FIB independently predicted non-thromb diseases were 0.726, 0.735, and 0.611, respectively. A new combined factor was generated by fitting age, D-dimer, and FIB with a logistic binary regression model. The AUC value of the combined prediction of non-thrombotic diseases was the maximum at 0.832, which had high diagnostic value, and its sensitivity and specificity were 0.572 and 0.070. CONCLUSION Elevated D-dimer levels were associated with age, increased FIB, and a variety of non-thrombotic diseases, and combination of age, D-dimer, and FIB had a certain predictive value for non-thrombotic diseases, but the combined model had a low specificity, other information needs to be combined in the clinic to improve diagnostic accuracy.
Humans ; Fibrin Fibrinogen Degradation Products ; Retrospective Studies ; Fibrinogen ; Predictive Value of Tests ; Thrombosis ; Age Factors ; ROC Curve ; Male ; Female ; Middle Aged ; Adult

OBJECTIVE: To investigate the predictive role of the modified Endothelial Activation and Stress Index (mEASIX) in the efficacy of chimeric antigen receptor T-cell (CAR-T) therapy and cytokine release syndrome (CRS). METHODS: The clinical data of 70 relapsed and refractory (R/R) B-cell tumor patients who were treated with CAR-T therapy from September 1, 2018 to February 28, 2023 in the Department of Hematology, Affiliated Hospital of Xuzhou Medical University, were retrospectively analyzed. The value of log-2 mEASIX before conditioning (-7 d) was calculated, and the patients were divided into a low-mEASIX group (42 patients) and a high-mEASIX group (28 patients) based on the cut-off value of 5.443 determined by the receiver operating characteristic (ROC) curve. Eventually, the predictive role of mEASIX before conditioning on the efficacy of CAR-T cell therapy and CRS was analyzed. RESULTS: The high-mEASIX group exhibited significantly worse median overall survival (OS) and median progression-free survival (PFS) in comparison to the low mEASIX group (OS: 3.2 months vs not reached, P < 0.01; PFS: 1.3 months vs 6.0 months, P =0.009). The incidence of grade ≥2 CRS in the high-mEASIX group was substantially higher than that in the low-mEASIX group (57.1% vs 19.0%, P =0.007). The degree of remission after CAR-T therapy (P =0.001), whether CRS occurs or not (P =0.041), the lactate dehydrogenase (LDH) level before conditioning (P =0.046), and the mEASIX score before conditioning (P =0.047) were independent influencing factors for the OS of patients receiving CAR-T cell therapy. CONCLUSION The mEASIX score before conditioning can predict OS and the incidence of grade ≥2 CRS in patients with relapsed and refractory B-cell tumors who receive CAR-T cell therapy.
Cytokine Release Syndrome/therapy* ; Immunotherapy, Adoptive/methods* ; Humans ; Lymphoma, B-Cell/therapy* ; Retrospective Studies ; Hematology ; China ; Receptors, Chimeric Antigen/blood* ; Predictive Value of Tests

Objective： The aim of this study is to explore the predictive value of biparametric magnetic resonance imaging(bpMRI)for pelvic lymph node metastasis in prostate cancer patients with PSA≤20 μg/L and establish a nomogram. Methods： The imaging data and clinical data of 363 patients undergoing radical prostatectomy and pelvic lymph node dissection in the First Affiliated Hospital of Nanjing Medical University from July 2018 to December 2023 were retrospectively analyzed. Univariate analysis and multivariate logistic regression were used to screen independent risk factors for pelvic lymph node metastasis in prostate cancer, and a nomogram of the clinical prediction model was established. Calibration curves were drawn to evaluate the accuracy of the model. Results： Multivariate logistic regression analysis showed extrocapusular extension (OR=8.08,95%CI=2.62－24.97, P<0.01), enlargement of pelvic lymph nodes (OR=4.45,95%CI=1.16－17.11,P=0.030), and biopsy ISUP grade(OR=1.97,95%CI=1.12－3.46, P=0.018)were independent risk factors for pelvic lymph node metastasis. The C-index of the prediction model was 0.834, which indicated that the model had a good prediction ability. The actual value of the model calibration curve and the prediction probability of the model fitted well, indicating that the model had a good accuracy. Further analysis of DCA curve showed that the model had good clinical application value when the risk threshold ranged from 0.05 to 0.70.Conclusion： For prostate cancer patients with PSA≤20 μg/L, bpMRI has a good predictive value for the pelvic lymph node metastasis of prostate cancer with extrocapusular extension, enlargement of pelvic lymph nodes and ISUP grade≥4.
Humans ; Male ; Prostatic Neoplasms/diagnostic imaging* ; Lymphatic Metastasis ; Retrospective Studies ; Nomograms ; Prostate-Specific Antigen/blood* ; Lymph Nodes/pathology* ; Pelvis ; Predictive Value of Tests ; Prostatectomy ; Lymph Node Excision ; Risk Factors ; Magnetic Resonance Imaging ; Logistic Models ; Middle Aged ; Aged