Support Vector Machine Research Articles

Statistical models versus machine learning approach for competing risks in proctological surgery.

Clinical risk prediction models are ubiquitous in many surgical domains. The traditional approach to develop these models involves the use of regression analysis. Machine learning algorithms are gaining in popularity as an alternative approach for prediction and classification problems. They can detect non-linear relationships between independent and dependent variables and incorporate many of them. In our work, we aimed to investigate the potential role of machine learning versus classical logistic regression for the preoperative risk assessment in proctological surgery. We used clinical data from a nationwide audit: the database consisted of 1510 patients affected by Goligher's grade III hemorrhoidal disease who underwent elective surgery. We collected anthropometric, clinical, and surgical data and we considered ten predictors to evaluate model-predictive performance. The clinical outcome was the complication rate evaluated at 30-day follow-up. Logistic regression and three machine learning techniques (Decision Tree, Support Vector Machine, Extreme Gradient Boosting) were compared in terms of area under the curve, balanced accuracy, sensitivity, and specificity. In our setting, machine learning and logistic regression models reached an equivalent predictive performance. Regarding the relative importance of the input features, all models agreed in identifying the most important factor. Combining and comparing statistical analysis and machine learning approaches in clinical field should be a common ambition, focused on improving and expanding interdisciplinary cooperation.

Sexual dimorphism of the humerus bones in a French sample: comparison of several statistical models including machine learning models.

Sex estimation is an important part of skeletal analysis and forensic identification. Traditionally pelvic traits are utilized for accurate sex estimation. However, the long bones, especially humerus, have been proved to be as effective for determine the sex of the individual.The aim of this study was to compare the predictive accuracy of seven statistical modelling techniques including classical statistical methods and machine learning algorithms, to assess the sexual dimorphism of humerus on a French sample based on a metric analysis of 26 measurements. A total of 98 humeral bones (divided in two samples) were measured. Seven statistical models were compared: Linear Discriminant Analysis (LDA), Regularized Discriminant Analysis (RDA), Penalized Logistic Regression (PLR), Flexible Discriminant Analysis (FDA), Support Vector Machine (SVM), and Artificial Neural Network (ANN) and Random Forest (RF).With cross validation, classification accuracy was greater than 90% (ranges between 92% and 98%) for all models without variable selection methods. The simplification of the models has improved the accuracy between 98% and 100% and also a reduction of the number of variables to 6 or less. Penalized logistic regression (PLR), Random Forest (RF) and Linear discriminant analysis (LDA) were the best accuracy models.The measurements made at the proximal part of the humerus (WTT, CSD), at distal part (BEW, WT, MAW, THT) and of the entire bone (PLCT) stand out among the various models.The present study suggests that the humerus is an interesting alternative for sex estimation and that non-classical statistical models can provide a new approach.

Detecting B-cell lymphoma-6 overexpression status in primary central nervous system lymphoma using multiparametric MRI-based machine learning.

In primary central nervous system lymphoma (PCNSL), B-cell lymphoma-6 (BCL-6) is an unfavorable prognostic biomarker. We aim to non-invasively detect BCL-6 overexpression in PCNSL patients using multiparametric MRI and machine learning techniques. 65 patients (101 lesions) with primary central nervous system lymphoma (PCNSL) diagnosed from January 2013 to July 2023, and all patients were randomly divided into a training set and a validation set according to a ratio of 8 to 2. ADC map derived from DWI (b = 0/1000 s/mm2), fast spin echo T2WI, T2FLAIR, were collected at 3.0 T. A total of 2234 radiomics features from the tumor segmentation area were extracted and LASSO were used to select features. Logistic regression (LR), Naive bayes (NB), Support vector machine (SVM), K-nearest Neighbor, (KNN) and Multilayer Perceptron (MLP), were used for machine learning, and sensitivity, specificity, accuracy F1-score, and area under the curve (AUC) was used to evaluate the detection performance of five classifiers, 6 groups with combinations of different sequences were fitted to 5 classifiers, and optimal classifier was obtained. BCL-6 status could be identified to varying degrees with 30 models based on radiomics, and model performance could be improved by combining different sequences and classifiers. Support vector machine (SVM) combined with three sequence group had the largest AUC (0.95) in training set and satisfactory AUC (0.87) in validation set. Multiparametric MRI based machine learning is promising in detecting BCL-6 overexpression.

Development and validation of a radiomics-visual evoked potential nomogram for preoperative prediction of visual outcome after endoscopic craniopharyngioma resection.

Craniopharyngiomas are rare, benign brain tumors that are primarily treated with surgery. Although the extended endoscopic endonasal approach (EEEA) has evolved as a more reliable surgical alternative and yields better visual outcomes than traditional craniotomy, postoperative visual deterioration remains one of the most common complications, and relevant risk factors are still poorly defined. Hence, identifying risk factors and developing a predictive model for postoperative visual deterioration is indeed necessary. However, there is still a lack of research on these topics. Therefore, the authors used the largest known case series of EEEA for craniopharyngioma to determine pertinent risk factors and develop a nomogram for the noninvasive preoperative prediction of visual outcome. A total of 483 cases of craniopharyngioma (338 in the training cohort, 145 in the validation cohort) between January 2019 and March 2023 were retrospectively reviewed, and related risk factors were identified. In total, 851 radiomic features from the MR images of each case were extracted. The least absolute shrinkage and selection operator algorithm was used to select features and construct the radiomic score (Rad-score). A support vector machine (SVM) classifier was adopted to construct a radiomic model. Moreover, a clinical-radiomic nomogram was built by multivariable logistic regression. The performance of the nomogram was assessed by its discrimination, calibration, and clinical utility. The overall incidence of postoperative visual deterioration was 9.1%. A lack of intraoperative visual evoked potential (VEP) monitoring (OR 0.221, p = 0.001), larger maximum tumor diameter (OR 1.052, p = 0.014), and tight adherence (OR 2.963, p = 0.044) were demonstrated as independent risk factors for postoperative visual deterioration. The radiomic model using the SVM based on 8 selected features exhibited good discrimination in predicting adhesion strength in the training and validation cohorts (area under the receiver operating characteristic curve [AUC] 0.85 vs 0.80). Moreover, the nomogram incorporating the Rad-score and clinical factors showed AUCs of 0.827 and 0.808 in the training and validation sets, respectively, fitting well in calibration curves. Decision curve analysis further confirmed the clinical usefulness of the nomogram. Intraoperative VEP monitoring was proven to help reduce postoperative visual deterioration, while tight adherence and larger maximum tumor diameter were confirmed as independent risk factors. The radiomic model allowed a noninvasive prediction of the adherence strength between the optic nerves and craniopharyngioma. The nomogram showed a promising performance for noninvasively predicting postoperative visual deterioration and may serve as a useful tool for clinical decision-making and patient counseling.

Predicting Polymerase Chain Reaction Success: Integrating the K-Word Order Model, Physicochemical Properties Modeling of Double Bases, and Support Vector Machine.

Polymerase Chain Reaction (PCR) has been a pivotal scientific technique since the twentieth century, and it is widely applied across various domains. Despite its ubiquity, challenges persist in efficiently amplifying specific DNA templates. While PCR experimental procedures have garnered significant attention, the analysis of the DNA template, which is the experiment's focal point, has been notably overlooked. This study addresses the uncertainty surrounding the amplification of DNA fragments using conventional Taq DNA polymerase-based PCR protocols. The imperative need to characterize DNA templates and devise a reliable method for predicting PCR success is underscored. In this study, we formulate a 72-dimensional feature vector representing a DNA template through the utilization of k-word order and modeling of physicochemical properties of double bases. Subsequently, a Support Vector Machine (SVM) model is employed to assess PCR results. A jackknife cross-validation test is used to evaluate the anticipated success rates, resulting in an overall accuracy of 95.77%. Sensitivity, specificity, and Matthew's Correlation Coefficient (MCC) stand at 95.75%, 95.79%, and 0.915, respectively.

Radiomics-Based Differentiation of Glioblastoma and Metastatic Disease: Impact of Different T1-Contrast-Enhanced Sequences on Radiomics Features and Model Performance.

Even though glioblastoma (GB) and brain metastases (BM) can be differentiated using radiomics, it remains unclear if the model performance may vary based on the contrast-enhanced sequence used. Our aim was to evaluate the radiomics-based model performance for differentiation between GB and brain metastases BM using MPRAGE and volumetric interpolated breath-hold examination (VIBE) T1-contrast-enhanced sequence. T1 contrast-enhanced (T1-CE) MPRAGE and VIBE sequences acquired in 108 patients (31 GBs and 77 BM) during the same MRI session were retrospectively evaluated. After standardized image preprocessing and segmentation, radiomics features were extracted from necrotic and enhancing tumor components. Pearson correlation analysis of radiomics features from tumor subcomponents was also performed. A total of 90 machine learning pipelines were evaluated using a 5-fold cross-validation. Performance was measured by mean area under the curve (AUC)-receiver operating characteristic (ROC), log loss, and Brier scores. A feature-wise comparison showed that the radiomics features between sequences were strongly correlated, with the highest correlation for shape-based features. The mean AUC across the top 10 pipelines ranged between 0.851 and 0.890 with T1-CE MPRAGE and between 0.869 and 0.907 with the T1-CE VIBE sequence. The top-performing models for the MPRAGE sequence commonly used support vector machines, while those for the VIBE sequence used either support vector machines or random forest. Common feature-reduction methods for top-performing models included linear combination filter and least absolute shrinkage and selection operator for both sequences. For the same machine learning feature-reduction pipeline, model performances were comparable (AUC-ROC difference range, -0.078-0.046). Radiomics features derived from T1-CE MPRAGE and VIBE sequences are strongly correlated and may have similar overall classification performance for differentiating GB from BM.

Assessing uncertainties in modeling the risk of geogenic groundwater contamination.

Assessing groundwater contamination risk is a critical aspect of preventing and managing groundwater pollution. There was a research gap in the investigation of uncertainties in modeling groundwater contamination risks in aquifers. This study addresses this gap using Bayesian Model Averaging (BMA), with a novel focus on risk exposures from geogenic contaminants, such as lead (Pb). This was achieved through the following methodology: (1) assessing aquifer vulnerability using the SPECTR framework; (2) generating a risk index for geogenic contaminants through unsupervised methods; (3) enhancing geogenic risk through three individual models, including Gene Expression Programming (GEP), M5P, and Support Vector Machines (SVM); (4) combining results from individual models using BMA; and (5) examining inherent uncertainties, accounting for both between-model and within-model variances. The model's efficacy was evaluated using measured Pb concentrations within the aquifer. The findings indicated that the unsupervised risk index had an acceptable correlation, while the individual models were accurate and enhanced the predictability of the data. BMA assigned the higher posterior probabilities (weight) to the SVM model, which indicates a positive correlation between the performance criteria of individual models and the weight values. Also, BMA revealed that the modeling uncertainty is influenced by within-model variance, primarily by the kriging interpolation method.

Integrating Retinal Segmentation Metrics with Machine Learning for Predictions from Mouse SD-OCT Scans

Purpose This study aimed to initially test whether machine learning approaches could categorically predict two simple biological features, mouse age and mouse species, using the retinal segmentation metrics. Methods The retinal layer thickness data obtained from C57BL/6 and DBA/2J mice were processed for machine learning after segmenting mouse retinal SD-OCT scans. Twenty-two models were trained to predict the mouse groups. The best neural network model was optimized for better outcomes. Prediction accuracy, the area under the curve, sensitivity, specificity, precision, and F-1 score values were obtained. Results The Wilcoxon Signed-Rank test provided significantly higher validation accuracy for neural networks than decision trees, discriminant analysis, support vector machines, and k-nearest neighbor classifiers (p = 0.005 for all). For C57BL/6-DBA/2J classification, a mean validation accuracy of 88.11 ± 3.92% (95% CI: 86.99-89.22) was achieved for the neural network when the optimized neural network had 92.31% final test accuracy with an area under the curve value of 0.9762, 94.44% sensitivity, 90.48% specificity, 89.47% precision, and 0.92 F-1 score. The optimized neural network model for age group differentiation had a final test accuracy of 82.05% with a 0.9064 area under the curve value, 77.27% sensitivity, 88.24% specificity, 89.47% precision, and 0.83 F-1 score. Conclusions These findings validate that machine learning, using segmentation metrics instead of images, can effectively analyze retinal OCT scans in mice for categorical predictions in experimental models. Expanding this approach with additional features, including histopathological and functional correlations, is expected to improve the prediction power further, promising valuable applications to predict more complex outcomes in experimental and clinical studies.

Advancing DNA Structural Analysis: A SERS Approach Free from Citrate Interference Combined with Machine Learning.

Surface-enhanced Raman spectroscopy (SERS) has become an indispensable tool for biomolecular analysis, yet the detection of DNA signals remains hindered by spectral interference from citrate ions, which overlap with key DNA features. This study introduces an innovative, ultrasensitive SERS platform utilizing thiol-modified silver nanoparticles (Ag@SDCNPs) that overcomes this challenge by eliminating citrate interference. This platform enables direct, interference-free detection and structural characterization of a wide range of DNA conformations, including single-stranded DNA (ssDNA), double-stranded DNA (dsDNA), i-motif, hairpin, G-quadruplex, and triple-stranded DNA (tsDNA). Employing calcium ions as aggregating agents and deuterated methanol as an internal standard, the system achieved high spectral quality and reproducibility. Machine learning (ML) techniques, such as linear discriminant analysis (LDA) and t-distributed stochastic neighbor embedding (t-SNE), were utilized for spectral classification, alongside support vector machines (SVM) for predictive modeling, yielding accuracies above 99%. These findings establish a robust and versatile platform for DNA structural analysis, offering transformative potential for applications in clinical diagnostics and biomedical research.

Spatial deconvolution from bulk DNA methylation profiles determines intratumoral epigenetic heterogeneity

BackgroundIntratumoral heterogeneity emerges from accumulating genetic and epigenetic changes during tumorigenesis, which may contribute to therapeutic failure and drug resistance. However, the lack of a quick and convenient approach to determine the intratumoral epigenetic heterogeneity (eITH) limit the application of eITH in clinical settings. Here, we aimed to develop a tool that can evaluate the eITH using the DNA methylation profiles from bulk tumors.MethodsGenomic DNA of three laser micro-dissected tumor regions, including digestive tract surface, central bulk, and invasive front, was extracted from formalin-fixed paraffin-embedded sections of colorectal cancer patients. The genome-wide methylation profiles were generated with methylation array. The most variable methylated probes were selected to construct a DNA methylation-based heterogeneity (MeHEG) estimation tool that can deconvolve the proportion of each reference tumor region with the support vector machine model-based method. A PCR-based assay for quantitative analysis of DNA methylation (QASM) was developed to specifically determine the methylation status of each CpG in MeHEG assay at single-base resolution to realize fast evaluation of epigenetic heterogeneity.ResultsIn the discovery set with 79 patients, the differentially methylated CpGs among the three tumor regions were found. The 7 most representative CpGs were identified and subsequently selected to develop the MeHEG algorithm. We validated its performance of deconvolution of tumor regions in an independent cohort. In addition, we showed the significant association of MeHEG-based epigenetic heterogeneity with the genomic heterogeneity in mutation and copy number variation in our in-house and TCGA cohorts. Besides, we found that the patients with higher MeHEG score had worse disease-free and overall survival outcomes. Finally, we found dynamic change of epigenetic heterogeneity based on MeHEG score in cancer cells under the treatment of therapeutic drugs.ConclusionBy developing a 7-loci panel using a machine learning approach combined with the QASM assay for PCR-based application, we present a valuable method for evaluating intratumoral heterogeneity. The MeHEG algorithm offers novel insights into tumor heterogeneity from an epigenetic perspective, potentially enriching current knowledge of tumor complexity and providing a new tool for clinical and research applications in cancer biology.

Combining a Risk Factor Score Designed From Electronic Health Records With a Digital Cytology Image Scoring System to Improve Bladder Cancer Detection: Proof-of-Concept Study.

To reduce the mortality related to bladder cancer, efforts need to be concentrated on early detection of the disease for more effective therapeutic intervention. Strong risk factors (eg, smoking status, age, professional exposure) have been identified, and some diagnostic tools (eg, by way of cystoscopy) have been proposed. However, to date, no fully satisfactory (noninvasive, inexpensive, high-performance) solution for widespread deployment has been proposed. Some new models based on cytology image classification were recently developed and bring good perspectives, but there are still avenues to explore to improve their performance. Our team aimed to evaluate the benefit of combining the reuse of massive clinical data to build a risk factor model and a digital cytology image-based model (VisioCyt) for bladder cancer detection. The first step relied on designing a predictive model based on clinical data (ie, risk factors identified in the literature) extracted from the clinical data warehouse of the Rennes Hospital and machine learning algorithms (logistic regression, random forest, and support vector machine). It provides a score corresponding to the risk of developing bladder cancer based on the patient's clinical profile. Second, we investigated 3 strategies (logistic regression, decision tree, and a custom strategy based on score interpretation) to combine the model's score with the score from an image-based model to produce a robust bladder cancer scoring system. We collected 2 data sets. The first set, including clinical data for 5422 patients extracted from the clinical data warehouse, was used to design the risk factor-based model. The second set was used to measure the models' performances and was composed of data for 620 patients from a clinical trial for which cytology images and clinicobiological features were collected. With this second data set, the combination of both models obtained areas under the curve of 0.82 on the training set and 0.83 on the test set, demonstrating the value of combining risk factor-based and image-based models. This combination offers a higher associated risk of cancer than VisioCyt alone for all classes, especially for low-grade bladder cancer. These results demonstrate the value of combining clinical and biological information, especially to improve detection of low-grade bladder cancer. Some improvements will need to be made to the automatic extraction of clinical features to make the risk factor-based model more robust. However, as of now, the results support the assumption that this type of approach will be of benefit to patients.

Machine Learning Analysis of Glutamate Receptor Activity in Developing Locus Coeruleus Neurons

: The developing brain undergoes a remarkable process of synapse production and maturation, particularly in glutamatergic synapses. In this study, we focused on the locus coeruleus (LC) nucleus, a brain region crucial for cognitive functions, to investigate the developmental changes in glutamatergic synaptic connections. Using the whole-cell patch clamp method, we recorded evoked excitatory postsynaptic currents (eEPSCs) from LC neurons in rats at ages 7, 14, and 21 days. To assess the developmental changes, we employed fractal analysis to compute fractal dimensions of AMPA and NMDA receptors, serving as markers for synaptic maturation. Our findings revealed a significant increase in fractal dimensions during the third postnatal week, indicating a developmental progression in the complexity and organization of synaptic connections. Pearson correlation analysis demonstrated a strong positive correlation between amplitude and fractal dimensions (p < 0.001), suggesting that the synaptic currents' amplitude is closely related to the fractal properties of the receptors. Furthermore, linear regression analysis showed a linear relationship between fractal dimensions and age (R2 = 0.843, p < 0.001), indicating that fractal analysis can be a robust tool for predicting age-dependent developmental changes. Additionally, we employed machine learning techniques to predict developmental changes based on AMPA and NMDA receptors. Support Vector Machine (SVM) models outperformed random forest models in accurately predicting age-dependent developmental changes, as indicated by the area under the curve (AUC) values. SVM achieved an AUC of 0.89 for AMPA receptors and 0.86 for NMDA receptors, demonstrating the effectiveness of fractal-based features in characterizing synaptic maturation. This study offers valuable insights into synaptic development in the LC nucleus and demonstrates the potential of fractal analysis as a tool to understand brain plasticity and early development. Fractal dimensions play a crucial role in characterizing the maturation of glutamatergic synapses, providing a foundation for further research in neural circuitry development.

Development of a machine learning tool to predict deep inspiration breath hold requirement for locoregional right-sided breast radiation therapy patients

Background and purpose. This study presents machine learning (ML) models that predict if deep inspiration breath hold (DIBH) is needed based on lung dose in right-sided breast cancer patients during the initial computed tomography (CT) appointment.Materials and methods. Anatomic distances were extracted from a single-institution dataset of free breathing (FB) CT scans from locoregional right-sided breast cancer patients. Models were developed using combinations of anatomic distances and ML classification algorithms (gradient boosting, k-nearest neighbors, logistic regression, random forest, and support vector machine) and optimized over 100 iterations using stratified 5-fold cross-validation. Models were grouped by the number of anatomic distances used during development; those with the highest validation accuracy were selected as final models. Final models were compared based on their predictive ability, measurement collection efficiency, and robustness to simulated user error during measurement collection.Results. This retrospective study included 238 patients treated between 2016 and 2021. Model development ended once eight anatomic distances were included, and the validation accuracy plateaued. The best performing model used logistic regression with four anatomic distances achieving 80.5% average testing accuracy, with minimal false negatives and positives (<27%). The anatomic distances required for prediction were collected within 3 min and were robust to simulated user error during measurement collection, changing accuracy by <5%.Conclusion. Our logistic regression model using four anatomic distances provided the best balance between efficiency, robustness, and ability to predict if DIBH was needed for locoregional right-sided breast cancer patients.

Ensemble machine learning models for lung cancer incidence risk prediction in the elderly: a retrospective longitudinal study

BackgroundIdentifying high risk factors and predicting lung cancer incidence risk are essential to prevention and intervention of lung cancer for the elderly. We aim to develop lung cancer incidence risk prediction model in the elderly to facilitate early intervention and prevention of lung cancer.MethodsWe stratified the population into six subgroups according to age and gender. For each subgroup, random forest, extreme gradient boosting, deep neural networks, support vector machine, multiple logistic regression and deep Q network (DQN) models were developed and validated. Models were trained and tested using samples from 2000 to 2015 and independent external validated through those from 2016 to 2019. The suitable model for lung cancer risk prediction and high risk factors identification was chosen based on internal validation and independent external validation.ResultsThe DQN model achieved the optimal prediction performance in stratified subgroups, with AUROC ranging from 0.937 to 0.953, recall ranging from 0.932 to 0.943, F2-score ranging from 0.929 to 0.946, precision ranging from 0.926 to 0.952, F1-score ranging from 0.933 to 0.963 and RMSE ranging from 0.21 to 0.27. SHAP values were supplied for model interpretability. High risk factors of lung cancer incidence were identified in the elderly. Men ≥ 65 carrying C > A/G > T mutation had the highest lung cancer incidence decrease of 39.5% after five years quitting in stratified elderly groups, which were 1.83 times more than women ≥ 65 not carrying C > A/G > T mutation.ConclusionsThe DQN model may be suitable for identifying high risk factors and predicting lung cancer risk with high performance. The proposed intervention and diagnosis pathways could be used for early screening and intervention before the occurrence of lung cancer, which could help oncologists develop targeted intervention strategies for the stratified elderly to reduce lung cancer incidence and improve therapeutic effect. Proposed method could also be used in predicting the risk of other chronic diseases to help conduct intervention and reduce incidence.

Novel Machine Learning Applications in Peripartum Care: A Scoping Review.

Machine learning (ML), a subtype of artificial intelligence (AI), presents predictive modeling and dynamic diagnostic tools to facilitate early interventions and improve decision-making. Considering the global challenges of maternal, fetal, and neonatal morbidity and mortality, ML holds the potential to enable significant improvements in maternal and neonatal health outcomes. We aimed to conduct a comprehensive review of ML applications in peripartum care, summarizing the potential of these tools to enhance clinical decision-making and identifying emerging trends and research gaps. We conducted a scoping review on MEDLINE, Cochrane Library, and EMBASE databases from inception to April 2024. We gathered additional relevant studies through snowball sampling. We meticulously screened titles and abstracts and chose full-text articles for further analysis. We included primary research articles and abstracts focusing on pregnant individuals, employing ML methods for peripartum care. No formal quality assessment was performed. Data were extracted using a custom template to capture study characteristics and ML models. Findings were synthesized using summary tables and figures to highlight key trends and results. Among 406 studies, 78% were published within the last five years. Most studies originated from high-income or well-resourced countries, with 27% from North America (including 24% from the United States) and 34% from Asia, predominantly China (18%). Studies from low- and middle-income regions were notably scarce, reflecting significant regional disparities. Predictive modeling tasks were the most prevalent (59%), followed by classification tasks (29%). Supervised learning dominated (90%), with algorithms such as Support Vector Machines, Random Forests, and Logistic Regression most commonly used. Key topics included fetal distress and acidemia (32%), preterm birth (22%), mode of delivery (13%), and birth weight (13%). Notably, Explainable AI methods were utilized in only 19% of studies, and external validation was performed in just 5%. Despite these advancements, only 1% of models resulted in accessible clinical tools, and none were fully integrated into healthcare systems. ML holds significant potential to enhance peripartum care by improving diagnostic accuracy and predictive capabilities. However, realizing this potential requires responsible AI practices, including robust validation with external datasets, prospective investigations across diverse populations, and the development of digital and data infrastructure for seamless integration into electronic health records. Additionally, transparent AI that provides insights into risk stratification logic is essential for clinician trust in ML tools. Future research should address understudied areas, prioritize neglected low-income settings, and explore advanced ML approaches to improve maternal and neonatal outcomes.

Classification of Molecular Subtypes of Breast Cancer Using Radiomic Features of Preoperative Ultrasound Images.

Radiomics has been used as a non-invasive medical image analysis technique for diagnosis and prognosis prediction of breast cancer. This study intended to use radiomics based on preoperative Doppler ultrasound images to classify four molecular subtypes of breast cancer. A total of 565 female breast cancer patients diagnosed by postoperative pathology in a hospital between 2014 and 2022 were included in this study. Radiomic features extracted from preoperative ultrasound images and clinical features were used to construct models for the classification of molecular subtypes of breast cancer. The least absolute shrinkage and selection operator (LASSO) regression was applied for the final screening of radiomic features and clinical features. Three classifiers including Logistic regression, support vector machine (SVM), and XGBoost were utilized to construct model. Model performance was assessed primarily by the area under the receiver operating characteristic curve (AUC) and 95% confidence interval (CI). The mean age of these patients was 54.58 (± 11.27) years. Of these 565 patients, 130 (23.01%) were Luminal A subtype, 329 (58.23%) were Luminal B subtype, 65 (11.50%) were human epidermal growth factor receptor-2 (HER-2) subtype, and 41 (7.26%) were triple negative (TN) subtype. A total of 12 clinical features and 8 radiomic features were selected for model construction. The AUC of the SVM model [0.826 (95%CI 0.808-0.845)] was higher than that of the Logistic regression model [0.776 (95%CI 0.756-0.796)] and the XGB model [0.800 (95%CI 0.779-0.821)] in the multiple classification of breast cancer. For the single classification of breast cancer, the AUC of the SVM model was 0.710 (95%CI 0.660-0.760) for Luminal A subtype, 0.639 (95%CI 0.592-0.685) for Luminal B subtype, 0.754 (95%CI 0.695-0.813) for HER-2 subtype, and 0.832 (95%CI 0.771-0.892) for TN subtype. The SVM model with radiomic features combined with clinical features shows good performance in classifying four molecular subtypes of breast cancer.

Gait patterns in unstable older patients related with vestibular hypofunction. Preliminary results in assessment with time-frequency analysis

Background Gait instability and falls significantly impact life quality and morbi-mortality in elderly populations. Early diagnosis of gait disorders is one of the most effective approaches to minimize severe injuries. Objective To find a gait instability pattern in older adults through an image representation of data collected by a single sensor. Methods A sample of 13 older adults (71-85 years old) with instability by vestibular hypofunction is compared to a sample of 19 adults (21-75 years old) without instability and normal vestibular function. Image representations of the gait signals acquired on a specific walk path were generated using a continuous wavelet transform and analyzed as a texture using grey level co-occurrence matrix metrics as features. A support vector machine (SVM) algorithm was used to discriminate subjects. Results First results show a good classification performance. According to analysis of extracted features, most information relevant to instability is concentrated in the medio-lateral acceleration (X axis) and the frontal plane angular rotation (Z axis gyroscope). Performing a ten-fold cross-validation through the first ten seconds of the sample dataset, the algorithm achieves a 92,3 F1 score corresponding to 12 true-positives, 1 false positive and 1 false negative. Discussion This preliminary report suggests that the method has potential use in assessing gait disorders in controlled and non-controlled environments. It suggests that deep learning methods could be explored given the availability of a larger population and data samples.

Development of a prediction model based on Hemoglobin, Albumin, Lymphocyte count, and Platelet-score for lymph node metastasis in rectal cancer.

This study aimed to evaluate the ability of the preoperative Hemoglobin, Albumin, Lymphocyte count, and Platelet (HALP) score to predict lymph node metastasis (LNM) in patients with rectal cancer (RC) and improve prediction accuracy by incorporating clinical parameters. Data from 263 patients with RC were analyzed. The receiver operating characteristic (ROC) curve was used to determine the optimal cutoff value (OCV) for the HALP score in predicting LNM. Based on this cutoff value, patients were divided into two groups. A baseline analysis was conducted to identify independent factors linked to LNM. A support vector machine (SVM) prediction model was developed, and its performance was evaluated using ROC, calibration curves, decision curve analysis, and Kolmogorov-Smirnov curve. The OCV for HALP score was 45.979. Patients were then classified into a low HALP group (n = 182) and a high HALP group (n = 81). The analysis found 21 clinical factors significantly associated with LNM. Among them, the key risk factors included high inflammatory status, poor nutritional condition, and a low HALP score. The SVM model incorporated these factors and showed robust predictive performance, with area under the curve values of 0.897, 0.813, and 0.750 for the training, validation, and testing datasets, respectively. The HALP score was significantly associated with LNM in RC patients. A machine learning model integrating the HALP score and inflammatory markers may be an effective tool for predicting LNM in RC.

Essential Oils as Antimicrobials against Acinetobacter baumannii: Experimental and Literature Data to Definite Predictive Quantitative Composition-Activity Relationship Models Using Machine Learning Algorithms.

Essential oils (EOs) exhibit a broad spectrum of biological activities; however, their clinical application is hindered by challenges, such as variability in chemical composition and chemical/physical instability. A critical limitation is the lack of chemical consistency across EO samples, which impedes standardization. Despite this, evidence suggests that EOs with differing chemical profiles often display similar (micro)biological activities, raising the possibility of standardizing EOs based on their biological effects rather than their chemical composition. This study explored the relationship between EO chemical composition and antibacterial activity against carbapenem-resistant Acinetobacter baumannii. A dataset comprising 82 EOs with known minimal inhibitory concentration values was compiled using both experimental results and literature data sourced from the AI4EssOil database (https://www.ai4essoil.com). Machine learning classification algorithms including Support Vector Machines, Random Forest, Gradient Boosting, Decision Trees, and K-Nearest Neighbors were employed to generate quantitative composition-activity relationship models. Model performance was assessed using internal and external prediction accuracy metrics with the Matthews correlation coefficient as the primary evaluation metrics. Features importance analysis, based on the Skater methodology, identified key chemical components influencing EO activity. The single chemical components limonene, eucalyptol, alpha-pinene, linalool, beta-caryophyllene, nerol, beta-pinene, neral, and carvacrol were highlighted as critical to biological efficacy. The predictive capacity of the ML models was validated against a test set of freshly extracted and chemically characterized EOs. The models demonstrated a 91% prediction accuracy for new EO samples, and a strong correlation was observed between predicted features importance and experimental inhibitory values for six selected pure compounds (limonene, eucalyptol, alpha-pinene, linalool, carvacrol, and thymol). Additionally, the machine learning approach was extended to cytotoxicity data from 3T3-Swiss fibroblasts for 61 EOs. The analysis revealed the potential to design EOs with both high antibacterial activity and low cytotoxicity through blending or selective enrichment with identified key components. These findings pave the way for biologically standardized EOs, enabling their rational design and optimization for clinical applications.

Redefining diagnostic lesional status in temporal lobe epilepsy with artificial intelligence.

Despite decades of advancements in diagnostic MRI, 30-50% of temporal lobe epilepsy (TLE) patients remain categorized as "non-lesional" (i.e., MRI negative or MRI-) based on visual assessment by human experts. MRI- patients face diagnostic uncertainty and significant delays in treatment planning. Quantitative MRI studies have demonstrated that MRI- patients often exhibit a TLE-specific pattern of temporal and limbic atrophy that may be too subtle for the human eye to detect. This signature pattern could be successfully translated into clinical use via artificial intelligence (AI) advances in computer-aided MRI interpretation, thereby improving the detection of brain "lesional" patterns associated with TLE. Here, we tested this hypothesis by employing a three-dimensional convolutional neural network (3D CNN) applied to a dataset of 1,178 scans from 12 different centers. 3D CNN was able to differentiate TLE from healthy controls with high accuracy (85.9% ± 2.8), significantly outperforming support vector machines based on hippocampal (74.4% ± 2.6) and whole-brain (78.3% ± 3.3) volumes. Our analysis subsequently focused on a subset of patients who achieved sustained seizure freedom post-surgery as a gold standard for confirming TLE. Importantly, MRI- patients from this cohort were accurately identified as TLE 82.7% ± 0.9 of the time, an encouraging finding since clinically these were all patients considered to be MRI- (i.e., not radiographically different than controls). The saliency maps from the CNN revealed that limbic structures, particularly medial temporal, cingulate, and orbitofrontal areas, were most influential in classification, confirming the importance of the well-established TLE signature atrophy pattern for diagnosis. Indeed, the saliency maps were similar in MRI+ and MRI- TLE groups, suggesting that even when humans cannot distinguish more subtle levels of atrophy, these MRI- patients are on the same continuum common across all TLE patients. As such, AI can identify TLE lesional patterns and AI-aided diagnosis has the potential to greatly enhance the neuroimaging diagnosis of TLE and redefine the concept of "lesional" TLE.