A machine learning model based on ultrasound image features to assess the risk of sentinel lymph node metastasis in breast cancer patients: Applications of scikit-learn and SHAP

Machine learning can reliably predict malignancy of breast lesions based on clinical and ultrasonographic features

PURPOSE

To establish a reliable machine learning model to predict malignancy in breast lesions identified by ultrasound (US) and optimize the negative predictive value to minimize unnecessary biopsies.

METHODS

We included clinical and ultrasonographic attributes from 1526 breast lesions classified as BI-RADS 3, 4a, 4b, 4c, 5, and 6 that underwent US-guided breast biopsy in four institutions. We selected the most informative attributes to train nine machine learning models, ensemble models and models with tuned threshold to make inferences about the diagnosis of BI-RADS 4a and 4b lesions (validation dataset). We tested the performance of the final model with 403 new suspicious lesions.

RESULTS

The most informative attributes were shape, margin, orientation and size of the lesions, the resistance index of the internal vessel, the age of the patient and the presence of a palpable lump. The highest mean negative predictive value (NPV) was achieved with the K-Nearest Neighbors algorithm (97.9%). Making ensembles did not improve the performance. Tuning the threshold did improve the performance of the models and we chose the algorithm XGBoost with the tuned threshold as the final one. The tested performance of the final model was: NPV 98.1%, false negative 1.9%, positive predictive value 77.1%, false positive 22.9%. Applying this final model, we would have missed 2 of the 231 malignant lesions of the test dataset (0.8%).

CONCLUSION

Machine learning can help physicians predict malignancy in suspicious breast lesions identified by the US. Our final model would be able to avoid 60.4% of the biopsies in benign lesions missing less than 1% of the cancer cases.

Gadoxetic acid-enhanced MRI for identifying cholangiocyte phenotype hepatocellular carcinoma by interpretable machine learning: individual application of SHAP

PURPOSE

Cholangiocyte phenotype hepatocellular carcinoma (HCC) is highly invasive. This study aims to develop and validate an optimal machine learning model to predict cholangiocyte phenotype HCC based on T1 mapping gadoxetic acid-enhanced MRI and to implement individual applications via the Shapley Additive explanation (SHAP).

METHODS

We included 180 patients with histologically confirmed HCC from two institutions. Clinical and MRI imaging features were screened for predicting cholangiocyte phenotype hepatocellular carcinoma using Least Absolute Shrinkage and Selection Operator (LASSO) and the logistic regression analysis. Five machine learning models were constructed based on these features. A Kaplan-Meier survival analysis aims to compare prognostic differences between cholangiocyte phenotype-positive HCC groups and classical (cholangiocyte phenotype-negative) HCC groups, and was conducted to explore the prognostic information of the optimal model.

RESULTS

The most significant clinicoradiological features, including the platelet-to-lymphocyte ratio (PLR), tumor capsule, target sign on hepatobiliary phase (HBP), and T1 relaxation time of 20 min (T1rt-20 min), were selected to construct the prediction model. Finally, we selected the eXtreme Gradient Boosting (XGBoost) model as the optimal predictive model, which achieved AUCs of 0.835, 0.830, 0.816 and 0.776 in training, internal validation, external validation, and prospective validation cohorts, respectively, for visual analysis via SHAP, in which T1rt-20 min made a significant contribution. Survival analysis showed a statistically significant difference in relapse-free survival (RFS) between cholangiocyte phenotype-positive HCC groups and classical HCC groups from institution I (hazard ratio [HR] 1.994; 95% CI, 1.059-3.758; P = 0.027), and the construction XGBoost model can be used to stratify RFS according to prognosis (HR, 1.986; 95% CI, 1.061-3.717; P = 0.029).

CONCLUSION

The machine learning model utilizing T1 mapping gadoxetic acid-enhanced MRI demonstrates significant potential in identifying cholangiocyte phenotype HCC. Furthermore, personalized prediction is enhanced through the application of SHAP, providing valuable insights to support clinical decision-making processes.

Enhanced breast cancer diagnosis using modified InceptionNet-V3: a deep learning approach for ultrasound image classification

Introduction

Breast cancer (BC) is a malignant neoplasm that originates in the mammary gland's cellular structures and remains one of the most prevalent cancers among women, ranking second in cancer-related mortality after lung cancer. Early and accurate diagnosis is crucial due to the heterogeneous nature of breast cancer and its rapid progression. However, manual detection and classification are often time-consuming and prone to errors, necessitating the development of automated and reliable diagnostic approaches.

Methods

Recent advancements in deep learning have significantly improved medical image analysis, demonstrating superior predictive performance in breast cancer detection using ultrasound images. Despite these advancements, training deep learning models from scratch can be computationally expensive and data-intensive. Transfer learning, leveraging pre-trained models on large-scale datasets, offers an effective solution to mitigate these challenges. In this study, we investigate and compare multiple deep-learning models for breast cancer classification using transfer learning. The evaluated architectures include modified InceptionV3, GoogLeNet, ShuffleNet, AlexNet, VGG-16, and SqueezeNet. Additionally, we propose a deep neural network model that integrates features from modified InceptionV3 to further enhance classification performance.

Results

The experimental results demonstrate that the modified InceptionV3 model achieves the highest classification accuracy of 99.10%, with a recall of 98.90%, precision of 99.00%, and an F1-score of 98.80%, outperforming all other evaluated models on the given datasets.

Discussion

The achieved findings underscore the potential of the proposed approach in enhancing diagnostic precision and confirm the superiority of the modified InceptionV3 model in breast cancer classification tasks.

Using weak signals to predict spontaneous breathing trial success: a machine learning approach

BACKGROUND

Weaning from mechanical ventilation (MV) is a key phase in the management of intensive care unit (ICU) patient. According to the WEAN SAFE study, weaning from MV initiation is defined as the first attempt to separate a patient from the ventilator and the success is the absence of reintubation (or death) within 7 days of extubation. Mortality rates increase with the difficulty of weaning, reaching 38% for the most challenging cases. Predicting the success of weaning is difficult, due to the complexity of factors involved. The many biosignals that are measured in patients during ventilation may be considered "weak signals", a concept rarely used in medicine. The aim of this research is to investigate the performance of machine learning (ML) models based on biosignals to predict spontaneous breathing trial success (SBT) using biosignals and to identify the most important variables.

METHODS

This retrospective study used data from two centers (Nice University Hospital, Archet and Pasteur) collected from 232 intensive care patients who underwent MV (149 successfully and 83 unsuccessfully) between January, 2020 and April, 2023. The study focuses on the development of ML algorithms to predict the success of the spontaneous breathing trial based on a combination of discrete variables and biosignals (time series) recorded during the 24 h prior to the SBT.

RESULTS

For the models tested, the best results were obtained with Support Vector Classifier model: AUC-PR 0.963 (0.936-0.970, p = 0.001), AUROC 0.922 (0.871-0.940, p < 0.001).

CONCLUSIONS

We found that ML models are effective in predicting the success of SBT based on biosignals. Predicting weaning from mechanical ventilation thus appears to be a promising area for the application of AI, through the development of multidimensional models to analyze weak signals.

Multiparametric MRI Radiomics With Machine Learning for Differentiating HER2-Zero, -Low, and -Positive Breast Cancer: Model Development, Testing, and Interpretability Analysis

BACKGROUND. MRI radiomics has been explored for three-tiered classification of HER2 expression levels (i.e., HER2-zero, HER2-low, or HER2-positive) in patients with breast cancer, although an understanding of how such models reach their predictions is lacking. OBJECTIVE. The purpose of this study was to develop and test multiparametric MRI radiomics machine learning models for differentiating three-tiered HER2 expression levels in patients with breast cancer, as well as to explain the contributions of model features through local and global interpretations with the use of Shapley additive explanation (SHAP) analysis. METHODS. This retrospective study included 737 patients (mean age, 54.1 ± 10.6 [SD] years) with breast cancer from two centers (center 1 [n = 578] and center 2 [n = 159]), all of whom underwent multiparametric breast MRI and had HER2 expression determined after excisional biopsy. Analysis entailed two tasks: differentiating HER2-negative (i.e., HER2-zero or HER2-low) tumors from HER2-positive tumors (task 1) and differentiating HER2-zero tumors from HER2-low tumors (task 2). For each task, patients from center 1 were randomly assigned in a 7:3 ratio to a training set (task 1: n = 405; task 2: n = 284) or an internal test set (task 1: n = 173; task 2: n = 122); patients from center 2 formed an external test set (task 1: n = 159; task 2: n = 105). Radiomic features were extracted from early phase dynamic contrast-enhanced (DCE) imaging, T2-weighted imaging, and DWI. For each task, a support vector machine (SVM) was used for feature selection, a multiparametric radiomics score (radscore) was computed using feature weights from SVM correlation coefficients, conventional MRI and combined models were constructed, and model performances were evaluated. SHAP analysis was used to provide local and global interpretations of the model outputs. RESULTS. In the external test set, for task 1, AUCs for the conventional MRI model, radscore, and the combined model were 0.624, 0.757, and 0.762, respectively; for task 2, the AUC for radscore was 0.754, and no conventional MRI model or combined model could be constructed. SHAP analysis identified early phase DCE imaging features as having the strongest influence for both tasks; T2-weighted imaging features also had a prominent role for task 2. CONCLUSION. The findings indicate suboptimal performance of MRI radiomics models for noninvasive characterization of HER2 expression. CLINICAL IMPACT. The study provides an example of the use of SHAP interpretation analysis to better understand predictions of imaging-based machine learning models.

Prediction of lateral lymph node metastasis with short diameter less than 8 mm in papillary thyroid carcinoma based on radiomics

OBJECTIVE

The aim of this study was to establish an ensemble learning model based on clinicopathological parameter and ultrasound radomics for assessing the risk of lateral cervical lymph node with short diameter less than 8 mm (small lymph nodes were used instead) metastasis in patients with papillary thyroid cancer (PTC), thereby guiding the selection of surgical methods.

METHODS

This retrospective analysis was conducted on 454 patients diagnosed with papillary thyroid carcinoma who underwent total thyroidectomy and lateral neck lymph node dissection or lymph node intraoperative frozen section biopsy at the First Hospital of China Medical University between January 2015 and April 2022. In a ratio of 8:2, 362(80%) patients were assigned to the training set and 92(20%) patients were assigned to the test set. Clinical pathological features and radomics features related to ultrasound imaging were extracted, followed by feature selection using recursive feature elimination (RFE). Based on distinct feature sets, we constructed ensemble learning models comprising random forest (RF), extreme gradient boosting (XGBoost), categorical boosting (CatBoost), gradient boosting decision tree (GBDT), and light gradient boosting machine (Lightgbm) to develop clinical models, radiomics models, and clinical-radiomic models. Through the comparison of performance metrics such as area under curve (AUC), accuracy (ACC), specificity (SPE), precision (PRE), recall rate, F1 score, mean squared error (MSE) etc., we identified the optimal model and visualized its results using shapley additive exPlanations (SHAP).

RESULTS

In this study, a total of 454 patients were included, among whom 342 PTC patients had small lymph node metastasis in the lateral neck region, while 112 did not have any metastasis. A total of 1035 features were initially considered for inclusion in this study, which were then narrowed down to 10 clinical features, 8 radiomics features, and 17 combined clinical-omics features. Based on these three feature sets, a total of fifteen ensemble learning models were established. In the test set, RF model in the clinical model is outperforms other models (AUC = 0.72, F1 = 0.75, Jaccard = 0.60 and Recall = 0.84), while CatBoost model in the radiomics model is superior to other models (AUC = 0.91, BA = 0.83 and SPE = 0.76). Among the clinical-radiomic models, Catboost exhibits optimal performance (AUC = 0.93, ACC = 0.88, BA = 0.87, F1 = 0.91, SPE = 0.83, PRE = 0.88, Jaccard = 0.83 and Recall = 0.92). Using the SHAP algorithm to visualize the operation process of the clinical-omics CatBoost model, we found that clinical omics features such as central lymph node metastasis (CLNM), Origin_Shape_Sphericity (o_shap_sphericity), LoG-sigma3_first order_ Skewness (log-3_fo_skewness), wavelet-HH_first order_Skewness (w-HH_fo_skewness) and wavelet-HH_first order_Skewness (sqr_gldm_DNUN) had the greatest impact on predicting the presence of lateral cervical small lymph node metastasis in PTC patients.

CONCLUSIONS

(1) In this study, among the ensemble learning models established based on clinicopathological features and radiomics features for predicting PTC lateral small lymph node metastasis, the clinical-radiomic CatBoost model has the best performance. (2) SHAP can visualize how the clinical and radiomics features affect the results and realize the interpretation of the model. (3) The combined CatBoost model can improve the diagnostic accuracy of suspicious lymph nodes with short diameter < 8 mm that are difficult to obtain accurate puncture results. The combined application of radiomics features is more accurate and reasonable than the prediction of clinical data alone, which helps to accurately evaluate the surgical scope and provide support for individual clinical decision making.

Machine learning-based prediction model of lower extremity deep vein thrombosis after stroke

This study aimed to apply machine learning (ML) techniques to develop and validate a risk prediction model for post-stroke lower extremity deep vein thrombosis (DVT) based on patients' limb function, activities of daily living (ADL), clinical laboratory indicators, and DVT preventive measures. We retrospectively analyzed 620 stroke patients. Eight ML models-logistic regression (LR), support vector machine (SVM), random forest (RF), decision tree (DT), neural network (NN), extreme gradient boosting (XGBoost), Bayesian (NB), and K-nearest neighbor (KNN)-were used to build the model. These models were extensively evaluated using ROC curves, AUC, PR curves, PRAUC, accuracy, sensitivity, specificity, and clinical decision curves (DCA). Shapley's additive explanation (SHAP) was used to determine feature importance. Finally, based on the optimal ML algorithm, different functional feature set models were compared with the Padua scale to select the best feature set model. Our results indicated that the RF algorithm demonstrated superior performance in various evaluation metrics, including AUC (0.74/0.73), PRAUC (0.58/0.58), accuracy (0.75/0.77), and sensitivity (0.78/0.80) in both the training set and test set. DCA analysis revealed that the RF model had the highest clinical net benefit. SHAP analysis showed that D-dimer had the most significant influence on DVT, followed by age, Brunnstrom stage (lower limb), prothrombin time (PT), and mobility ability. The RF algorithm can predict post-stroke DVT to guide clinical practice.

Predicting Sudden Sensorineural Hearing Loss Recovery with Patient-Personalized Seigel's Criteria Using Machine Learning

BACKGROUND

Accurate prognostic prediction is crucial for managing Idiopathic Sudden Sensorineural Hearing Loss (ISSHL). Previous studies developing ISSHL prognosis models often overlooked individual variability in hearing damage by relying on fixed frequency domains. This study aims to develop models predicting ISSHL prognosis one month after treatment, focusing on patient-specific hearing impairments.

METHODS

Patient-Personalized Seigel's Criteria (PPSC) were developed considering patient-specific hearing impairment related to ISSHL criteria. We performed a statistical test to assess the shift in the recovery assessment when applying PPSC. The utilized dataset of 581 patients comprised demographic information, health records, laboratory testing, onset and treatment, and hearing levels. To reduce the model's reliance on hearing level features, we used only the averages of hearing levels of the impaired frequencies. Then, model development, evaluation, and interpretation proceeded.

RESULTS

The chi-square test (p-value: 0.106) indicated that the shift in recovery assessment is not statistically significant. The soft-voting ensemble model was most effective, achieving an Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.864 (95% CI: 0.801-0.927), with model interpretation based on the SHapley Additive exPlanations value.

CONCLUSIONS

With PPSC, providing a hearing assessment comparable to traditional Seigel's criteria, the developed models successfully predicted ISSHL recovery one month post-treatment by considering patient-specific impairments.

Interpretable machine learning model integrating clinical and elastosonographic features to detect renal fibrosis in Asian patients with chronic kidney disease

BACKGROUND

Non-invasive renal fibrosis assessment is critical for tailoring personalized decision-making and managing follow-up in patients with chronic kidney disease (CKD). We aimed to exploit machine learning algorithms using clinical and elastosonographic features to distinguish moderate-severe fibrosis from mild fibrosis among CKD patients.

METHODS

A total of 162 patients with CKD who underwent shear wave elastography examinations and renal biopsies at our institution were prospectively enrolled. Four classifiers using machine learning algorithms, including eXtreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), Light Gradient Boosting Machine (LightGBM), and K-Nearest Neighbor (KNN), which integrated elastosonographic features and clinical characteristics, were established to differentiate moderate-severe renal fibrosis from mild forms. The area under the receiver operating characteristic curve (AUC) and average precision were employed to compare the performance of constructed models, and the SHapley Additive exPlanations (SHAP) strategy was used to visualize and interpret the model output.

RESULTS

The XGBoost model outperformed the other developed machine learning models, demonstrating optimal diagnostic performance in both the primary (AUC = 0.97, 95% confidence level (CI) 0.94-0.99; average precision = 0.97, 95% CI 0.97-0.98) and five-fold cross-validation (AUC = 0.85, 95% CI 0.73-0.98; average precision = 0.90, 95% CI 0.86-0.93) datasets. The SHAP approach provided visual interpretation for XGBoost, highlighting the features' impact on the diagnostic process, wherein the estimated glomerular filtration rate provided the largest contribution to the model output, followed by the elastic modulus, then renal length, renal resistive index, and hypertension.

CONCLUSION

This study proposed an XGBoost model for distinguishing moderate-severe renal fibrosis from mild forms in CKD patients, which could be used to assist clinicians in decision-making and follow-up strategies. Moreover, the SHAP algorithm makes it feasible to visualize and interpret the feature processing and diagnostic processes of the model output.

Deep Learning Prediction of Axillary Lymph Node Metastasis in Breast Cancer Patients Using Clinical Implication-Applied Preprocessed CT Images

Background: Accurate detection of axillary lymph node (ALN) metastases in breast cancer is crucial for clinical staging and treatment planning. This study aims to develop a deep learning model using clinical implication-applied preprocessed computed tomography (CT) images to enhance the prediction of ALN metastasis in breast cancer patients. Methods: A total of 1128 axial CT images of ALN (538 malignant and 590 benign lymph nodes) were collected from 523 breast cancer patients who underwent preoperative CT scans between January 2012 and July 2022 at Hallym University Medical Center. To develop an optimal deep learning model for distinguishing metastatic ALN from benign ALN, a CT image preprocessing protocol with clinical implications and two different cropping methods (fixed size crop [FSC] method and adjustable square crop [ASC] method) were employed. The images were analyzed using three different convolutional neural network (CNN) architectures (ResNet, DenseNet, and EfficientNet). Ensemble methods involving and combining the selection of the two best-performing CNN architectures from each cropping method were applied to generate the final result. Results: For the two different cropping methods, DenseNet consistently outperformed ResNet and EfficientNet. The area under the receiver operating characteristic curve (AUROC) for DenseNet, using the FSC and ASC methods, was 0.934 and 0.939, respectively. The ensemble model, which combines the performance of the DenseNet121 architecture for both cropping methods, delivered outstanding results with an AUROC of 0.968, an accuracy of 0.938, a sensitivity of 0.980, and a specificity of 0.903. Furthermore, distinct trends observed in gradient-weighted class activation mapping images with the two cropping methods suggest that our deep learning model not only evaluates the lymph node itself, but also distinguishes subtler changes in lymph node margin and adjacent soft tissue, which often elude human interpretation. Conclusions: This research demonstrates the promising performance of a deep learning model in accurately detecting malignant ALNs in breast cancer patients using CT images. The integration of clinical considerations into image processing and the utilization of ensemble methods further improved diagnostic precision.

Machine learning and new insights for breast cancer diagnosis

Breast cancer (BC) is the most prominent form of cancer among females all over the world. The current methods of BC detection include X-ray mammography, ultrasound, computed tomography, magnetic resonance imaging, positron emission tomography and breast thermographic techniques. More recently, machine learning (ML) tools have been increasingly employed in diagnostic medicine for its high efficiency in detection and intervention. The subsequent imaging features and mathematical analyses can then be used to generate ML models, which stratify, differentiate and detect benign and malignant breast lesions. Given its marked advantages, radiomics is a frequently used tool in recent research and clinics. Artificial neural networks and deep learning (DL) are novel forms of ML that evaluate data using computer simulation of the human brain. DL directly processes unstructured information, such as images, sounds and language, and performs precise clinical image stratification, medical record analyses and tumour diagnosis. Herein, this review thoroughly summarizes prior investigations on the application of medical images for the detection and intervention of BC using radiomics, namely DL and ML. The aim was to provide guidance to scientists regarding the use of artificial intelligence and ML in research and the clinic.

Application of artificial intelligence in ultrasound imaging for predicting lymph node metastasis in breast cancer: A meta-analysis

BACKGROUND

This study aims to comprehensively evaluate the accuracy and effectiveness of ultrasound imaging based on artificial intelligence algorithms in predicting lymph node metastasis in breast cancer patients through a meta-analysis.

METHODS

We systematically searched PubMed, Embase, and Cochrane Library for literature published up to May 2023. The search terms included artificial intelligence, ultrasound, breast cancer, and lymph node. Studies meeting the inclusion criteria were selected, and data were extracted for analysis. The main evaluation indicators included sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and area under the curve (AUC). The heterogeneity was assessed using the Cochrane Q test combined with the I^2 statistic expressing the percentage of total effect variation that can be attributed to the effect variation between studies, as recommended by the Cochrane Handbook for heterogeneity quantification. A threshold p-value of 0.10 was considered to compensate for the low power of the Q test. Sensitivity analysis was performed to assess the stability of individual studies, and publication bias was determined with funnel plots. Additionally, fagan plots were used to assess clinical utility.

RESULTS

Ten studies involving 4726 breast cancer patients were included in the meta-analysis. The results showed that ultrasound imaging based on artificial intelligence algorithms had high accuracy and effectiveness in predicting lymph node metastasis in breast cancer patients. The pooled sensitivity was 0.88 (95% CI: 0.81-0.93; P < 0.001; I2 = 84.68), specificity was 0.75 (95% CI: 0.66-0.83; P < 0.001; I2 = 91.11), and AUC was 0.89 (95% CI: 0.86-0.91). The positive likelihood ratio was 3.5 (95% CI: 2.6-4.8), negative likelihood ratio was 0.16 (95% CI: 0.10-0.26), and diagnostic odds ratio was 23 (95% CI: 13-40). However, the combined sensitivity of ultrasound imaging based on non-AI algorithms for predicting lymph node metastasis in breast cancer patients was 0.78 (95%CI: 0.63-0.88), the specificity was 0.76 (95%CI: 0.63-0.86), and the AUC was 0.84 (95%CI: 0.80-0.87). The positive likelihood ratio was 3.3 (95% CI: 1.9-5.6), the negative likelihood ratio was 0.29 (95% CI: 0.15-0.54), and the diagnostic odds ratio was 11 (95% CI: 4-33). Due to limited sample size (n = 2), meta-analysis was not conducted for the outcome of predicting lymph node metastasis burden.

CONCLUSION

Ultrasound imaging based on artificial intelligence algorithms holds promise in predicting lymph node metastasis in breast cancer patients, demonstrating high accuracy and effectiveness. The application of this technology helps in the diagnosis and treatment decisions for breast cancer patients and is expected to become an important tool in future clinical practice.

Differentiation Between Phyllodes Tumor and Fibroadenoma of the Breast: A Radiomics Prediction Model Based on Full-Field Digital Mammography & Digital Tomosynthesis

OBJECTIVE

To assess the diagnostic performance of FFDM-based and DBT-based radiomics models to differentiate breast phyllodes tumors from fibroadenomas.

METHODS

192 patients (93 phyllodes tumors and 99 fibroadenomas) who underwent mammography were retrospectively enrolled. Radiomic features were respectively extracted from FFDM and the clearest slice of DBT images. A least absolute shrinkage and selection operator (LASSO) regression was used to select radiomics features. A combined model was constructed by radiomics and radiological signatures. Machine learning classification was done using logistic regression based on radiomics or radiological signatures (clinical model). Four radiologists were tested on phyllodes tumors and fibroadenomas with and without optimal model assistance. The area under the receiver operating characteristic (ROC) curve (AUC) was computed to assess the performance of each model or radiologist. The Delong test and McNemar's test were performed to compare the performance.

RESULTS

The combined model yielded the highest performance with an AUC of 0.948 (95%CI: 0.889-1.000) in the testing set, slightly higher than the FFDM-radiomics model (AUC of 0.937, 95%CI: 0.841-0.984) and the DBT-radiomics model (AUC of 0.860, 95%CI: 0.742-0.936) and significantly superior to the clinical model (AUC of 0.719, 95%CI: 0.585-0.829). With the combined model aid, the AUCs of four radiologists were improved from 0.808 to 0.914 (p=0.079), 0.759 to 0.888 (p=0.015), 0.717 to 0.846 (p=0.004), and 0.629 to 0.803 (p=0.001).

CONCLUSION

Radiomics analysis based on FFDM and DBT shows promise in differentiating phyllodes tumors from fibroadenomas.

Predictive value for axillary lymph node metastases in early breast cancer: Based on contrast-enhanced ultrasound characteristics of the primary lesion and sentinel lymph node

OBJECTIVE

To evaluate the value of contrast-enhanced ultrasound (CEUS) characteristics based on primary lesion combined with lymphatic contrast-enhanced ultrasound (LCEUS) patterns of SLN in predicting axillary lymph node metastasis (ALNM) with T1-2N0 breast cancer.

METHODS

A retrospective study was conducted in 118 patients with clinically confirmed T1-2N0 breast cancer. Conventional ultrasound (CUS) and CEUS characteristics of the primary lesion and enhancement patterns of SLN were recorded. The risk factors associated with ALNM were selected by univariate and binary logistic regression analysis, and the receiver operating characteristic (ROC) curve was drawn for the evaluation of predictive ALNM metastasis performance.

RESULTS

Univariate analysis showed that age, HER-2 status, tumor size, nutrient vessels, extended range of enhancement lesion, and the enhancement patterns of SLN were significant predictive features of ALNM. Further binary logistic regression analysis indicated that the extended range of enhancement lesion (p <  0.001) and the enhancement patterns of SLN (p <  0.001) were independent risk factors for ALNM. ROC analysis showed that the AUC of the combination of these two indicators for predicting ALNM was 0.931 (95% CI: 0.887-0.976, sensitivity: 75.0%, specificity: 99.8%).

CONCLUSION

The CEUS characteristics of primary lesion combined with enhancement patterns of SLN are highly valuable in predicting ALNM and can guide clinical axillary surgery decision-making in early breast cancer.

Prognostic models for breast cancer: based on logistics regression and Hybrid Bayesian Network

BACKGROUND

To construct two prognostic models to predict survival in breast cancer patients; to compare the efficacy of the two models in the whole group and the advanced human epidermal growth factor receptor-2-positive (HER2+) subgroup of patients; to conclude whether the Hybrid Bayesian Network (HBN) model outperformed the logistics regression (LR) model.

METHODS

In this paper, breast cancer patient data were collected from the SEER database. Data processing and analysis were performed using Rstudio 4.2.0, including data preprocessing, model construction and validation. The L_DVBN algorithm in Julia0.4.7 and bnlearn package in R was used to build and evaluate the HBN model. Data with a diagnosis time of 2018(n = 23,384) were distributed randomly as training and testing sets in the ratio of 7:3 using the leave-out method for model construction and internal validation. External validation of the model was done using the dataset of 2019(n = 8128). Finally, the late HER2 + patients(n = 395) was selected for subgroup analysis. Accuracy, calibration and net benefit of clinical decision making were evaluated for both models.

RESULTS

The HBN model showed that seventeen variables were associated with survival outcome, including age, tumor size, site, histologic type, radiotherapy, surgery, chemotherapy, distant metastasis, subtype, clinical stage, ER receptor, PR receptor, clinical grade, race, marital status, tumor laterality, and lymph node. The AUCs for the internal validation of the LR and HBN models were 0.831 and 0.900; The AUCs for the external validation of the LR and HBN models on the whole population were 0.786 and 0.871; the AUCs for the external validation of the two models on the subgroup population were 0.601 and 0.813.

CONCLUSION

The accuracy, net clinical benefit, and calibration of the HBN model were better than LR model. The predictive efficacy of both models decreased and the difference was greater in advanced HER2 + patients, which means the HBN model had higher robustness and more stable predictive performance in the subgroup.

Development of Machine Learning-Based Web System for Estimating Pleural Effusion Using Multi-Frequency Bioelectrical Impedance Analyses

BACKGROUND

Transthoracic impedance values have not been widely used to measure extravascular pulmonary water content due to accuracy and complexity concerns. Our aim was to develop a foundational model for a novel system aiming to non-invasively estimate the intrathoracic condition of heart failure patients.

METHODS

We employed multi-frequency bioelectrical impedance analysis to simultaneously measure multiple frequencies, collecting electrical, physical, and hematological data from 63 hospitalized heart failure patients and 82 healthy volunteers. Measurements were taken upon admission and after treatment, and longitudinal analysis was conducted.

RESULTS

Using a light gradient boosting machine, and a decision tree-based machine learning method, we developed an intrathoracic estimation model based on electrical measurements and clinical findings. Out of the 286 features collected, the model utilized 16 features. Notably, the developed model demonstrated high accuracy in discriminating patients with pleural effusion, achieving an area under the receiver characteristic curves (AUC) of 0.905 (95% CI: 0.870-0.940, p < 0.0001) in the cross-validation test. The accuracy significantly outperformed the conventional frequency-based method with an AUC of 0.740 (95% CI: 0.688-0.792, and p < 0.0001).

CONCLUSIONS

Our findings indicate the potential of machine learning and transthoracic impedance measurements for estimating pleural effusion. By incorporating noninvasive and easily obtainable clinical and laboratory findings, this approach offers an effective means of assessing intrathoracic conditions.

Predicting the Activity of Oral Lichen Planus with Glycolysis-related Molecules: A Scikit-learn-based Function

OBJECTIVE

Oral lichen planus (OLP) is one of the most common oral mucosa diseases, and is mainly mediated by T lymphocytes. The metabolic reprogramming of activated T cells has been shown to transform from oxidative phosphorylation to aerobic glycolysis. The present study investigated the serum levels of glycolysis-related molecules (lactate dehydrogenase, LDH; pyruvic acid, PA; lactic acid, LAC) in OLP, and the correlation with OLP activity was assessed using the reticular, atrophic and erosive lesion (RAE) scoring system.

METHODS

Univariate and multivariate linear regression functions based on scikit-learn were designed to predict the RAE scores in OLP patients, and the performance of these two machine learning functions was compared.

RESULTS

The results revealed that the serum levels of PA and LAC were upregulated in erosive OLP (EOLP) patients, when compared to healthy volunteers. Furthermore, the LDH and LAC levels were significantly higher in the EOLP group than in the nonerosive OLP (NEOLP) group. All glycolysis-related molecules were positively correlated to the RAE scores. Among these, LAC had a strong correlation. The univariate function that involved the LAC level and the multivariate function that involved all glycolysis-related molecules presented comparable prediction accuracy and stability, but the latter was more time-consuming.

CONCLUSION

It can be concluded that the serum LAC level can be a user-friendly biomarker to monitor the OLP activity, based on the univariate function developed in the present study. The intervention of the glycolytic pathway may provide a potential therapeutic strategy.

A Hybrid Algorithm of ML and XAI to Prevent Breast Cancer: A Strategy to Support Decision Making

Worldwide, the coronavirus has intensified the management problems of health services, significantly harming patients. Some of the most affected processes have been cancer patients' prevention, diagnosis, and treatment. Breast cancer is the most affected, with more than 20 million cases and at least 10 million deaths by 2020. Various studies have been carried out to support the management of this disease globally. This paper presents a decision support strategy for health teams based on machine learning (ML) tools and explainability algorithms (XAI). The main methodological contributions are: first, the evaluation of different ML algorithms that allow classifying patients with and without cancer from the available dataset; and second, an ML methodology mixed with an XAI algorithm, which makes it possible to predict the disease and interpret the variables and how they affect the health of patients. The results show that first, the XGBoost Algorithm has a better predictive capacity, with an accuracy of 0.813 for the train data and 0.81 for the test data; and second, with the SHAP algorithm, it is possible to know the relevant variables and their level of significance in the prediction, and to quantify the impact on the clinical condition of the patients, which will allow health teams to offer early and personalized alerts for each patient.

The prediction of distant metastasis risk for male breast cancer patients based on an interpretable machine learning model

OBJECTIVES

This research was designed to compare the ability of different machine learning (ML) models and nomogram to predict distant metastasis in male breast cancer (MBC) patients and to interpret the optimal ML model by SHapley Additive exPlanations (SHAP) framework.

METHODS

Four powerful ML models were developed using data from male breast cancer (MBC) patients in the SEER database between 2010 and 2015 and MBC patients from our hospital between 2010 and 2020. The area under curve (AUC) and Brier score were used to assess the capacity of different models. The Delong test was applied to compare the performance of the models. Univariable and multivariable analysis were conducted using logistic regression.

RESULTS

Of 2351 patients were analyzed; 168 (7.1%) had distant metastasis (M1); 117 (5.0%) had bone metastasis, and 71 (3.0%) had lung metastasis. The median age at diagnosis is 68.0 years old. Most patients did not receive radiotherapy (1723, 73.3%) or chemotherapy (1447, 61.5%). The XGB model was the best ML model for predicting M1 in MBC patients. It showed the largest AUC value in the tenfold cross validation (AUC:0.884; SD:0.02), training (AUC:0.907; 95% CI: 0.899-0.917), testing (AUC:0.827; 95% CI: 0.802-0.857) and external validation (AUC:0.754; 95% CI: 0.739-0.771) sets. It also showed powerful ability in the prediction of bone metastasis (AUC: 0.880, 95% CI: 0.856-0.903 in the training set; AUC: 0.823, 95% CI:0.790-0.848 in the test set; AUC: 0.747, 95% CI: 0.727-0.764 in the external validation set) and lung metastasis (AUC: 0.906, 95% CI: 0.877-0.928 in training set; AUC: 0.859, 95% CI: 0.816-0.891 in the test set; AUC: 0.756, 95% CI: 0.732-0.777 in the external validation set). The AUC value of the XGB model was larger than that of nomogram in the training (0.907 vs 0.802) and external validation (0.754 vs 0.706) sets.

CONCLUSIONS

The XGB model is a better predictor of distant metastasis among MBC patients than other ML models and nomogram; furthermore, the XGB model is a powerful model for predicting bone and lung metastasis. Combining with SHAP values, it could help doctors intuitively understand the impact of each variable on outcome.

The Role of AI in Breast Cancer Lymph Node Classification: A Comprehensive Review

Breast cancer is a significant health issue affecting women worldwide, and accurately detecting lymph node metastasis is critical in determining treatment and prognosis. While traditional diagnostic methods have limitations and complications, artificial intelligence (AI) techniques such as machine learning (ML) and deep learning (DL) offer promising solutions for improving and supplementing diagnostic procedures. Current research has explored state-of-the-art DL models for breast cancer lymph node classification from radiological images, achieving high performances (AUC: 0.71-0.99). AI models trained on clinicopathological features also show promise in predicting metastasis status (AUC: 0.74-0.77), whereas multimodal (radiomics + clinicopathological features) models combine the best from both approaches and also achieve good results (AUC: 0.82-0.94). Once properly validated, such models could greatly improve cancer care, especially in areas with limited medical resources. This comprehensive review aims to compile knowledge about state-of-the-art AI models used for breast cancer lymph node metastasis detection, discusses proper validation techniques and potential pitfalls and limitations, and presents future directions and best practices to achieve high usability in real-world clinical settings.

Radiomics model of diffusion-weighted whole-body imaging with background signal suppression (DWIBS) for predicting axillary lymph node status in breast cancer

BACKGROUND

In breast cancer diagnosis and treatment, non-invasive prediction of axillary lymph node (ALN) metastasis can help avoid complications related to sentinel lymph node biopsy.

OBJECTIVE

This study aims to develop and evaluate machine learning models using radiomics features extracted from diffusion-weighted whole-body imaging with background signal suppression (DWIBS) examination for predicting the ALN status.

METHODS

A total of 100 patients with histologically proven, invasive, clinically N0 breast cancer who underwent DWIBS examination consisting of short tau inversion recovery (STIR) and DWIBS sequences before surgery were enrolled. Radiomic features were calculated using segmented primary lesions in DWIBS and STIR sequences and were divided into training (n = 75) and test (n = 25) datasets based on the examination date. Using the training dataset, optimal feature selection was performed using the least absolute shrinkage and selection operator algorithm, and the logistic regression model and support vector machine (SVM) classifier model were constructed with DWIBS, STIR, or a combination of DWIBS and STIR sequences to predict ALN status. Receiver operating characteristic curves were used to assess the prediction performance of radiomics models.

RESULTS

For the test dataset, the logistic regression model using DWIBS, STIR, and a combination of both sequences yielded an area under the curve (AUC) of 0.765 (95% confidence interval: 0.548-0.982), 0.801 (0.597-1.000), and 0.779 (0.567-0.992), respectively, whereas the SVM classifier model using DWIBS, STIR, and a combination of both sequences yielded an AUC of 0.765 (0.548-0.982), 0.757 (0.538-0.977), and 0.779 (0.567-0.992), respectively.

CONCLUSIONS

Use of machine learning models incorporating with the quantitative radiomic features derived from the DWIBS and STIR sequences can potentially predict ALN status.

Gradient Boosting Machine Identified Predictive Variables for Breast Cancer Patients Pre- and Post-Radiotherapy: Preliminary Results of an 8-Year Follow-Up Study

Radiotherapy (RT) is part of the standard treatment of breast cancer (BC) because of its effects on relapse reduction and survival. However, response to treatment is highly variable, and some patients may develop disease progression (DP), a second primary cancer, or may succumb to the disease. Antioxidant systems and inflammatory processes are associated with the onset and development of BC and play a role in resistance to treatment. Here, we report our investigation into the clinical evolution of BC patients, and the impact of RT on the circulating levels of the antioxidant enzyme paraoxonase-1 (PON1), cytokines, and other standard biochemical and hematological variables. Gradient Boosting Machine (GBM) algorithm was used to identify predictive variables. This was a retrospective study in 237 patients with BC. Blood samples were obtained pre- and post-RT, with samples of healthy women used as control subjects. Results showed that 24 patients had DP eight years post-RT, and eight patients developed a second primary tumor. The algorithm identified interleukin-4 and total lymphocyte counts as the most relevant indices discriminating between BC patients and control subjects, while neutrophils, total leukocytes, eosinophils, very low-density lipoprotein cholesterol, and PON1 activity were potential predictors of fatal outcome.