Feature selection for outcome prediction in oesophageal cancer using genetic algorithm and random forest classifier

Comparative study of machine learning approaches integrated with genetic algorithm for IVF success prediction

INTRODUCTION

IVF is a widely-used assisted reproductive technology with a consistent success rate of around 30%, and improving this rate is crucial due to emotional, financial, and health-related implications for infertile couples. This study aimed to develop a model for predicting IVF outcome by comparing five machine-learning techniques.

METHOD

The research approached five prominent machine learning algorithms, including Random Forest, Artificial Neural Network (ANN), Support Vector Machine (SVM), Recursive Partitioning and Regression Trees (RPART), and AdaBoost, in the context of IVF success prediction. The study also incorporated GA as a feature selection method to enhance the predictive models' robustness.

RESULTS

Findings demonstrate that AdaBoost, particularly when combined with GA feature selection, achieved the highest accuracy rate of 89.8%. Using GA, Random Forest also demonstrated strong performance, achieving an accuracy rate of 87.4%. Genetic Algorithm significantly improved the performance of all classifiers, emphasizing the importance of feature selection. Ten crucial features, including female age, AMH, endometrial thickness, sperm count, and various indicators of oocyte and embryo quality, were identified as key determinants of IVF success.

CONCLUSION

These findings underscore the potential of machine learning and feature selection techniques to assist IVF clinicians in providing more accurate predictions, enabling tailored treatment plans for each patient. Future research and validation can further enhance the practicality and reliability of these predictive models in clinical IVF practice.

Computed tomography-based radiomics nomogram for prediction of lympho-vascular and perineural invasion in esophageal squamous cell cancer patients: a retrospective cohort study

PURPOSE

Lympho-vascular invasion (LVI) and perineural invasion (PNI) have been established as prognostic factors in various types of cancers. The preoperative prediction of LVI and PNI has the potential to guide personalized medicine strategies for patients with esophageal squamous cell cancer (ESCC). This study investigates whether radiomics features derived from preoperative contrast-enhanced CT could predict LVI and PNI in ESCC patients.

METHODS AND MATERIALS

A retrospective cohort of 544 ESCC patients who underwent esophagectomy were included in this study. Preoperative contrast-enhanced CT images, pathological results of PNI and LVI, and clinical characteristics were collected. For each patient, the gross tumor volume (GTV-T) and lymph nodes volume (GTV-N) were delineated and four categories of radiomics features (first-order, shape, textural and wavelet) were extracted from GTV-T and GTV-N. The Mann-Whitney U test was used to select significant features associated with LVI and PNI in turn. Subsequently, radiomics signatures for LVI and PNI were constructed using LASSO regression with ten-fold cross-validation. Significant clinical characteristics were combined with radiomics signature to develop two nomogram models for predicting LVI and PNI, respectively. The area under the curve (AUC) and calibration curve were used to evaluate the predictive performance of the models.

RESULTS

The radiomics signature for LVI prediction consisted of 28 features, while the PNI radiomics signature comprised 14 features. The AUCs of the LVI radiomics signature were 0.77 and 0.74 in the training and validation groups, respectively, while the AUCs of the PNI radiomics signature were 0.69 and 0.68 in the training and validation groups. The nomograms incorporating radiomics signatures and significant clinical characteristics such as age, gender, thrombin time and D-Dimer showed improved predictive performance for both LVI (AUC: 0.82 and 0.80 in the training and validation group) and PNI (AUC: 0.75 and 0.72 in the training and validation groups) compared to the radiomics signature alone.

CONCLUSION

The radiomics features extracted from preoperative contrast-enhanced CT of gross tumor and lymph nodes have demonstrated their potential in predicting LVI and PNI in ESCC patients. Furthermore, the incorporation of clinical characteristics has shown additional value, resulting in improved predictive performance.

Artificial intelligence methods available for cancer research

Cancer is a heterogeneous and multifaceted disease with a significant global footprint. Despite substantial technological advancements for battling cancer, early diagnosis and selection of effective treatment remains a challenge. With the convenience of large-scale datasets including multiple levels of data, new bioinformatic tools are needed to transform this wealth of information into clinically useful decision-support tools. In this field, artificial intelligence (AI) technologies with their highly diverse applications are rapidly gaining ground. Machine learning methods, such as Bayesian networks, support vector machines, decision trees, random forests, gradient boosting, and K-nearest neighbors, including neural network models like deep learning, have proven valuable in predictive, prognostic, and diagnostic studies. Researchers have recently employed large language models to tackle new dimensions of problems. However, leveraging the opportunity to utilize AI in clinical settings will require surpassing significant obstacles-a major issue is the lack of use of the available reporting guidelines obstructing the reproducibility of published studies. In this review, we discuss the applications of AI methods and explore their benefits and limitations. We summarize the available guidelines for AI in healthcare and highlight the potential role and impact of AI models on future directions in cancer research.

Baseline CT radiomics features to predict pathological complete response of advanced esophageal squamous cell carcinoma treated with neoadjuvant chemotherapy using paclitaxel and cisplatin

PURPOSE

To develop a CT radiomics model to predict pathological complete response (pCR) of advanced esophageal squamous cell carcinoma (ESCC) toneoadjuvant chemotherapy using paclitaxel and cisplatin.

MATERIALS AND METHODS

326 consecutive patients with advanced ESCC from two hospitals undergoing baseline contrast-enhanced CT followed by neoadjuvant chemotherapy using paclitaxel and cisplatin were enrolled, including 115 patients achieving pCR and 211 patients without pCR. Of the 271 cases from 1st hospital, 188 and 83 cases were randomly allocated to the training and test cohorts, respectively. The 55 patients from a second hospital were assigned as an external validation cohort. Region of interest was segmented on the baseline thoracic contrast-enhanced CT. Useful radiomics features were generated by dimension reduction using least absolute shrinkage and selection operator. The optimal radiomics features were chosen using support vector machine (SVM). Discriminating performance was assessed with area under the receiver operating characteristic curve (ROC) and F-1score. The calibration curves and Brier score were used to evaluate the predictive accuracy.

RESULTS

Eight radiomics features were selected to create radiomics models related to pCR of advanced ESCC (P-values < 0.01 for both the training and test cohorts). SVM model showed the best performance (AUCs = 0.929, 0.868 and 0.866, F-1scores = 0.857, 0.847 and 0.737 in the training, test and external validation cohorts, respectively). The calibration curves and Brier scores indicated goodness-of-fit and its great predictive accuracy.

CONCLUSION

CT radiomics models could well help predict pCR of advanced ESCC, and SVM model could be a suitable predictive model.

Plasma glycoproteomics delivers high-specificity disease biomarkers by detecting site-specific glycosylation abnormalities

INTRODUCTION

The human plasma glycoproteome holds enormous potential to identify personalized biomarkers for diagnostics. Glycoproteomics has matured into a technology for plasma N-glycoproteome analysis but further evolution towards clinical applications depends on the clinical validity and understanding of protein- and site-specific glycosylation changes in disease.

OBJECTIVES

Here, we exploited the uniqueness of a patient cohort of genetic defects in well-defined glycosylation pathways to assess the clinical applicability of plasma N-glycoproteomics.

METHODS

Comparative glycoproteomics was performed of blood plasma from 40 controls and 74 patients with 13 different genetic diseases that impact the protein N-glycosylation pathway. Baseline glycosylation in healthy individuals was compared to reference glycome and intact transferrin protein mass spectrometry data. Use of glycoproteomics data for biomarker discovery and sample stratification was evaluated by multivariate chemometrics and supervised machine learning. Clinical relevance of site-specific glycosylation changes were evaluated in the context of genetic defects that lead to distinct accumulation or loss of specific glycans. Integrated analysis of site-specific glycoproteome changes in disease was performed using chord diagrams and correlated with intact transferrin protein mass spectrometry data.

RESULTS

Glycoproteomics identified 191 unique glycoforms from 58 unique peptide sequences of 34 plasma glycoproteins that span over 3 magnitudes of abundance in plasma. Chemometrics identified high-specificity biomarker signatures for each of the individual genetic defects with better stratification performance than the current diagnostic standard method. Bioinformatic analyses revealed site-specific glycosylation differences that could be explained by underlying glycobiology and protein-intrinsic factors.

CONCLUSION

Our work illustrates the strong potential of plasma glycoproteomics to significantly increase specificity of glycoprotein biomarkers with direct insights in site-specific glycosylation changes to better understand the glycobiological mechanisms underlying human disease.

A radiomics and genomics-derived model for predicting metastasis and prognosis in colorectal cancer

Approximately 50% of colorectal cancer (CRC) patients would develop metastasis with poor prognosis, therefore, it is necessary to effectively predict metastasis in clinical treatment. In this study, we aimed to establish a machine-learning model for predicting metastasis in CRC patients by considering radiomics and transcriptomics simultaneously. Here, 1023 patients with CRC from three centers were collected and divided into five queues (Dazhou Central Hospital n = 517, Nanchong Central Hospital n = 120 and the Cancer Genome Atlas (TCGA) n = 386). A total of 854 radiomics features were extracted from tumor lesions on CT images, and 217 differentially expressed genes were obtained from non-metastasis and metastasis tumor tissues using RNA sequencing. Based on radiotranscriptomic (RT) analysis, a novel RT model was developed and verified through genetic algorithms (GA). Interleukin (IL)-26, a biomarker in RT model, was verified for its biological function in CRC metastasis. Furthermore, 15 radiomics variables were screened through stepwise regression, which was highly correlated with the IL26 expression level. Finally, a radiomics model (RA) was established by combining GA and stepwise regression analysis with radiomics features. The RA model exhibited favorable discriminatory ability and accuracy for metastasis prediction in two independent verification cohorts. We designed multicenter, multi-scale cohorts to construct and verify novel combined radiomics and genomics models for predicting metastasis in CRC. Overall, RT model and RA model might help clinicians in directing personalized diagnosis and therapeutic regimen selection for patients with CRC.

External Validation of Robust Radiomic Signature to Predict 2-Year Overall Survival in Non-Small-Cell Lung Cancer

Lung cancer is the second most fatal disease worldwide. In the last few years, radiomics is being explored to develop prediction models for various clinical endpoints in lung cancer. However, the robustness of radiomic features is under question and has been identified as one of the roadblocks in the implementation of a radiomic-based prediction model in the clinic. Many past studies have suggested identifying the robust radiomic feature to develop a prediction model. In our earlier study, we identified robust radiomic features for prediction model development. The objective of this study was to develop and validate the robust radiomic signatures for predicting 2-year overall survival in non-small cell lung cancer (NSCLC). This retrospective study included a cohort of 300 stage I-IV NSCLC patients. Institutional 200 patients' data were included for training and internal validation and 100 patients' data from The Cancer Image Archive (TCIA) open-source image repository for external validation. Radiomic features were extracted from the CT images of both cohorts. The feature selection was performed using hierarchical clustering, a Chi-squared test, and recursive feature elimination (RFE). In total, six prediction models were developed using random forest (RF-Model-O, RF-Model-B), gradient boosting (GB-Model-O, GB-Model-B), and support vector(SV-Model-O, SV-Model-B) classifiers to predict 2-year overall survival (OS) on original data as well as balanced data. Model validation was performed using 10-fold cross-validation, internal validation, and external validation. Using a multistep feature selection method, the overall top 10 features were chosen. On internal validation, the two random forest models (RF-Model-O, RF-Model-B) displayed the highest accuracy; their scores on the original and balanced datasets were 0.81 and 0.77 respectively. During external validation, both the random forest models' accuracy was 0.68. In our study, robust radiomic features showed promising predictive performance to predict 2-year overall survival in NSCLC.

Trustworthy Data and AI Environments for Clinical Prediction: Application to Crisis-Risk in People With Depression

Depression is a common mental health condition that often occurs in association with other chronic illnesses, and varies considerably in severity. Electronic Health Records (EHRs) contain rich information about a patient's medical history and can be used to train, test and maintain predictive models to support and improve patient care. This work evaluated the feasibility of implementing an environment for predicting mental health crisis among people living with depression based on both structured and unstructured EHRs. A large EHR from a mental health provider, Mersey Care, was pseudonymised and ingested into the Natural Language Processing (NLP) platform CogStack, allowing text content in binary clinical notes to be extracted. All unstructured clinical notes and summaries were semantically annotated by MedCAT and BioYODIE NLP services. Cases of crisis in patients with depression were then identified. Random forest models, gradient boosting trees, and Long Short-Term Memory (LSTM) networks, with varying feature arrangement, were trained to predict the occurrence of crisis. The results showed that all the prediction models can use a combination of structured and unstructured EHR information to predict crisis in patients with depression with good and useful accuracy. The LSTM network that was trained on a modified dataset with only 1000 most-important features from the random forest model with temporality showed the best performance with a mean AUC of 0.901 and a standard deviation of 0.006 using a training dataset and a mean AUC of 0.810 and 0.01 using a hold-out test dataset. Comparing the results from the technical evaluation with the views of psychiatrists shows that there are now opportunities to refine and integrate such prediction models into pragmatic point-of-care clinical decision support tools for supporting mental healthcare delivery.

The gap before real clinical application of imaging-based machine-learning and radiomic models for chemoradiation outcome prediction in esophageal cancer: a systematic review and meta-analysis

BACKGROUND

Due to tumoral heterogeneity and the lack of robust biomarkers, the prediction of chemoradiotherapy response and prognosis in patients with esophageal cancer (EC) is challenging. The goal of this study was to assess the study quality and clinical value of machine learning and radiomic-based quantitative imaging studies for predicting the outcomes of EC patients after chemoradiotherapy.

MATERIALS AND METHODS

PubMed, Embase, and Cochrane were searched for eligible articles. The methodological quality and risk of bias were evaluated using the Radiomics Quality Score (RQS), Image Biomarkers Standardization Initiative (IBSI) Guideline, and Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD) statement, as well as the modified Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. A meta-analysis of the evidence focusing on predicting chemoradiotherapy response and outcome in EC patients was implemented.

RESULTS

Forty-six studies were eligible for qualitative synthesis. The mean RQS score was 9.07, with an adherence rate of 42.52%. The adherence rates of the TRIPOD and IBSI were 61.70 and 43.17%, respectively. Ultimately, 24 studies were included in the meta-analysis, of which 16 studies had a pooled sensitivity, specificity, and area under the curve (AUC) of 0.83 (0.76-0.89), 0.83 (0.79-0.86), and 0.84 (0.81-0.87) in neoadjuvant chemoradiotherapy datasets, as well as 0.84 (0.75-0.93), 0.89 (0.83-0.93), and 0.93 (0.90-0.95) in definitive chemoradiotherapy datasets, respectively. Moreover, radiomics could distinguish patients from the low-risk and high-risk groups with different disease-free survival (DFS) (pooled hazard ratio: 3.43, 95% CI 2.39-4.92) and overall survival (pooled hazard ratio: 2.49, 95% CI 1.91-3.25). The results of subgroup and regression analyses showed that some of the heterogeneity was explained by the combination with clinical factors, sample size, and usage of the deep learning (DL) signature.

CONCLUSIONS

Noninvasive radiomics offers promising potential for optimizing treatment decision-making in EC patients. However, it is necessary to make scientific advancements in EC radiomics regarding reproducibility, clinical usefulness analysis, and open science categories. Improved model reporting of study objectives, blind assessment, and image processing steps are required to help promote real clinical applications of radiomics in EC research.

Unipolar and Bipolar Depression Detection and Classification Based on Actigraphic Registration of Motor Activity Using Machine Learning and Uniform Manifold Approximation and Projection Methods

Modern technology frequently uses wearable sensors to monitor many aspects of human behavior. Since continuous records of heart rate and activity levels are typically gathered, the data generated by these devices have a lot of promise beyond counting the number of daily steps or calories expended. Due to the patient's inability to obtain the necessary information to understand their conditions and detect illness, such as depression, objectively, methods for evaluating various mental disorders, such as the Montgomery-Asberg depression rating scale (MADRS) and observations, currently require a significant amount of effort on the part of specialists. In this study, a novel dataset was provided, comprising sensor data gathered from depressed patients. The dataset included 32 healthy controls and 23 unipolar and bipolar depressive patients with motor activity recordings. Along with the sensor data collected over several days of continuous measurement for each patient, some demographic information was also offered. The result of the experiment showed that less than 70 of the 100 epochs of the model's training were completed. The Cohen Kappa score did not even pass 0.1 in the validation set, due to an imbalance in the class distribution, whereas in the second experiment, the majority of scores peaked in about 20 epochs, but because training continued during each epoch, it took much longer for the loss to decline before it fell below 0.1. In the second experiment, the model soon reached an accuracy of 0.991, which is as expected given the outcome of the UMAP dimensionality reduction. In the last experiment, UMAP and neural networks worked together to produce the best outcomes. They used a variety of machine learning classification algorithms, including the nearest neighbors, linear kernel SVM, Gaussian process, and random forest. This paper used the UMAP unsupervised machine learning dimensionality reduction without the neural network and showed a slightly lower score (QDA). By considering the ratings of the patient's depressive symptoms that were completed by medical specialists, it is possible to better understand the relationship between depression and motor activity.

Comprehensive evaluation of machine learning algorithms for predicting sleep-wake conditions and differentiating between the wake conditions before and after sleep during pregnancy based on heart rate variability

Introduction

Perinatal women tend to have difficulties with sleep along with autonomic characteristics. This study aimed to identify a machine learning algorithm capable of achieving high accuracy in predicting sleep-wake conditions and differentiating between the wake conditions before and after sleep during pregnancy based on heart rate variability (HRV).

Methods

Nine HRV indicators (features) and sleep-wake conditions of 154 pregnant women were measured for 1 week, from the 23rd to the 32nd weeks of pregnancy. Ten machine learning and three deep learning methods were applied to predict three types of sleep-wake conditions (wake, shallow sleep, and deep sleep). In addition, the prediction of four conditions, in which the wake conditions before and after sleep were differentiated-shallow sleep, deep sleep, and the two types of wake conditions-was also tested.

Results and Discussion

In the test for predicting three types of sleep-wake conditions, most of the algorithms, except for Naïve Bayes, showed higher areas under the curve (AUCs; 0.82-0.88) and accuracy (0.78-0.81). The test using four types of sleep-wake conditions with differentiation between the wake conditions before and after sleep also resulted in successful prediction by the gated recurrent unit with the highest AUC (0.86) and accuracy (0.79). Among the nine features, seven made major contributions to predicting sleep-wake conditions. Among the seven features, "the number of interval differences of successive RR intervals greater than 50 ms (NN50)" and "the proportion dividing NN50 by the total number of RR intervals (pNN50)" were useful to predict sleep-wake conditions unique to pregnancy. These findings suggest alterations in the vagal tone system specific to pregnancy.

Performance of radiomics-based artificial intelligence systems in the diagnosis and prediction of treatment response and survival in esophageal cancer: a systematic review and meta-analysis of diagnostic accuracy

Radiomics can interpret radiological images with more detail and in less time compared to the human eye. Some challenges in managing esophageal cancer can be addressed by incorporating radiomics into image interpretation, treatment planning, and predicting response and survival. This systematic review and meta-analysis provides a summary of the evidence of radiomics in esophageal cancer. The systematic review was carried out using Pubmed, MEDLINE, and Ovid EMBASE databases-articles describing radiomics in esophageal cancer were included. A meta-analysis was also performed; 50 studies were included. For the assessment of treatment response using 18F-FDG PET/computed tomography (CT) scans, seven studies (443 patients) were included in the meta-analysis. The pooled sensitivity and specificity were 86.5% (81.1-90.6) and 87.1% (78.0-92.8). For the assessment of treatment response using CT scans, five studies (625 patients) were included in the meta-analysis, with a pooled sensitivity and specificity of 86.7% (81.4-90.7) and 76.1% (69.9-81.4). The remaining 37 studies formed the qualitative review, discussing radiomics in diagnosis, radiotherapy planning, and survival prediction. This review explores the wide-ranging possibilities of radiomics in esophageal cancer management. The sensitivities of 18F-FDG PET/CT scans and CT scans are comparable, but 18F-FDG PET/CT scans have improved specificity for AI-based prediction of treatment response. Models integrating clinical and radiomic features facilitate diagnosis and survival prediction. More research is required into comparing models and conducting large-scale studies to build a robust evidence base.

Prognostic value of pre-therapeutic FDG-PET radiomic analysis in gastro-esophageal junction cancer

The main aim of this study was to evaluate the prognostic value of radiomic approach in pre-therapeutic ¹⁸F-fluorodeoxyglucose positron-emission tomography (FDG-PET/CT) in a large cohort of patients with gastro-esophageal junction cancer (GEJC). This was a retrospective monocenter study including 97 consecutive patients with GEJC who underwent a pre-therapeutic FDG-PET and were followed up for 3 years. Standard first-order radiomic PET indices including SUV_max, SUV_mean, SUV_peak, MTV and TLG and 32 textural features (TFs) were calculated using LIFEx software on PET imaging. Prognostic significance of these parameters was assessed in univariate and multivariate analysis. Relapse-free survival (RFS) and overall survival (OS) were respectively chosen as primary and secondary endpoints. An internal validation cohort was used by randomly drawing one-third of included patients. The main characteristics of this cohort were: median age of 65 years [41-88], sex ratio H/F = 83/14, 81.5% of patients with a histopathology of adenocarcinoma and 43.3% with a stage IV disease. The median follow-up was 28.5 months [4.2-108.5]. Seventy-seven (79.4%) patients had locoregional or distant progression or recurrence and 71 (73.2%) died. In univariate analysis, SUV_mean, Histogram-Entropy and 2 TFs (GLCM-Homogeneity and GLCM-Energy) were significantly correlated with RFS and OS, as well as 2 others TFs (GLRLM-LRE and GLRLM-GLNU) with OS only. In multivariate analysis, Histogram-Entropy remained an independent prognostic factor of both RFS and OS whereas SUV_mean was an independent prognostic factor of OS only. These results were partially confirmed in our internal validation cohort of 33 patients. Our results suggest that radiomic approach reveals independent prognostic factors for survival in patients with GEJC.

Optimizing prognostic factors of five-year survival in gastric cancer patients using feature selection techniques with machine learning algorithms: a comparative study

BACKGROUND

Gastric cancer is the most common malignant tumor worldwide and a leading cause of cancer deaths. This neoplasm has a poor prognosis and heterogeneous outcomes. Survivability prediction may help select the best treatment plan based on an individual's prognosis. Numerous clinical and pathological features are generally used in predicting gastric cancer survival, and their influence on the survival of this cancer has not been fully elucidated. Moreover, the five-year survivability prognosis performances of feature selection methods with machine learning (ML) classifiers for gastric cancer have not been fully benchmarked. Therefore, we adopted several well-known feature selection methods and ML classifiers together to determine the best-paired feature selection-classifier for this purpose.

METHODS

This was a retrospective study on a dataset of 974 patients diagnosed with gastric cancer in the Ayatollah Talleghani Hospital, Abadan, Iran. First, four feature selection algorithms, including Relief, Boruta, least absolute shrinkage and selection operator (LASSO), and minimum redundancy maximum relevance (mRMR) were used to select a set of relevant features that are very informative for five-year survival prediction in gastric cancer patients. Then, each feature set was fed to three classifiers: XG Boost (XGB), hist gradient boosting (HGB), and support vector machine (SVM) to develop predictive models. Finally, paired feature selection-classifier methods were evaluated to select the best-paired method using the area under the curve (AUC), accuracy, sensitivity, specificity, and f1-score metrics.

RESULTS

The LASSO feature selection algorithm combined with the XG Boost classifier achieved an accuracy of 89.10%, a specificity of 87.15%, a sensitivity of 89.42%, an AUC of 89.37%, and an f1-score of 90.8%. Tumor stage, history of other cancers, lymphatic invasion, tumor site, type of treatment, body weight, histological type, and addiction were identified as the most significant factors affecting gastric cancer survival.

CONCLUSIONS

This study proved the worth of the paired feature selection-classifier to identify the best path that could augment the five-year survival prediction in gastric cancer patients. Our results were better than those of previous studies, both in terms of the time required to form the models and the performance measurement criteria of the algorithms. These findings may be very promising and can, therefore, inform clinical decision-making and shed light on future studies.

Machine Learning Analysis of RNA-seq Data for Diagnostic and Prognostic Prediction of Colon Cancer

Data from omics studies have been used for prediction and classification of various diseases in biomedical and bioinformatics research. In recent years, Machine Learning (ML) algorithms have been used in many different fields related to healthcare systems, especially for disease prediction and classification tasks. Integration of molecular omics data with ML algorithms has offered a great opportunity to evaluate clinical data. RNA sequence (RNA-seq) analysis has been emerged as the gold standard for transcriptomics analysis. Currently, it is being used widely in clinical research. In our present work, RNA-seq data of extracellular vesicles (EV) from healthy and colon cancer patients are analyzed. Our aim is to develop models for prediction and classification of colon cancer stages. Five different canonical ML and Deep Learning (DL) classifiers are used to predict colon cancer of an individual with processed RNA-seq data. The classes of data are formed on the basis of both colon cancer stages and cancer presence (healthy or cancer). The canonical ML classifiers, which are k-Nearest Neighbor (kNN), Logistic Model Tree (LMT), Random Tree (RT), Random Committee (RC), and Random Forest (RF), are tested with both forms of the data. In addition, to compare the performance with canonical ML models, One-Dimensional Convolutional Neural Network (1-D CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM) DL models are utilized. Hyper-parameter optimizations of DL models are constructed by using genetic meta-heuristic optimization algorithm (GA). The best accuracy in cancer prediction is obtained with RC, LMT, and RF canonical ML algorithms as 97.33%. However, RT and kNN show 95.33% performance. The best accuracy in cancer stage classification is achieved with RF as 97.33%. This result is followed by LMT, RC, kNN, and RT with 96.33%, 96%, 94.66%, and 94%, respectively. According to the results of the experiments with DL algorithms, the best accuracy in cancer prediction is obtained with 1-D CNN as 97.67%. BiLSTM and LSTM show 94.33% and 93.67% performance, respectively. In classification of the cancer stages, the best accuracy is achieved with BiLSTM as 98%. 1-D CNN and LSTM show 97% and 94.33% performance, respectively. The results reveal that both canonical ML and DL models may outperform each other for different numbers of features.

Machine learning-based prediction of 1-year mortality in hypertensive patients undergoing coronary revascularization surgery

BACKGROUND

Machine learning (ML) has shown promising results in all fields of medicine, including preventive cardiology. Hypertensive patients are at higher risk of mortality after coronary artery bypass graft (CABG) surgery; thus, we aimed to design and evaluate five ML models to predict 1-year mortality among hypertensive patients who underwent CABG.

HYOTHESIS

ML algorithms can significantly improve mortality prediction after CABG.

METHODS

Tehran Heart Center's CABG data registry was used to extract several baseline and peri-procedural characteristics and mortality data. The best features were chosen using random forest (RF) feature selection algorithm. Five ML models were developed to predict 1-year mortality: logistic regression (LR), RF, artificial neural network (ANN), extreme gradient boosting (XGB), and naïve Bayes (NB). The area under the curve (AUC), sensitivity, and specificity were used to evaluate the models.

RESULTS

Among the 8,493 hypertensive patients who underwent CABG (mean age of 68.27 ± 9.27 years), 303 died in the first year. Eleven features were selected as the best predictors, among which total ventilation hours and ejection fraction were the leading ones. LR showed the best prediction ability with an AUC of 0.82, while the least AUC was for the NB model (0.79). Among the subgroups, the highest AUC for LR model was for two age range groups (50-59 and 80-89 years), overweight, diabetic, and smoker subgroups of hypertensive patients.

CONCLUSIONS

All ML models had excellent performance in predicting 1-year mortality among CABG hypertension patients, while LR was the best regarding AUC. These models can help clinicians assess the risk of mortality in specific subgroups at higher risk (such as hypertensive ones).

Deep learning-based important weights-only transfer learning approach for COVID-19 CT-scan classification

COVID-19 has become a pandemic for the entire world, and it has significantly affected the world economy. The importance of early detection and treatment of the infection cannot be overstated. The traditional diagnosis techniques take more time in detecting the infection. Although, numerous deep learning-based automated solutions have recently been developed in this regard, nevertheless, the limitation of computational and battery power in resource-constrained devices makes it difficult to deploy trained models for real-time inference. In this paper, to detect the presence of COVID-19 in CT-scan images, an important weights-only transfer learning method has been proposed for devices with limited runt-time resources. In the proposed method, the pre-trained models are made point-of-care devices friendly by pruning less important weight parameters of the model. The experiments were performed on two popular VGG16 and ResNet34 models and the empirical results showed that pruned ResNet34 model achieved 95.47% accuracy, 0.9216 sensitivity, 0.9567 F-score, and 0.9942 specificity with 41.96% fewer FLOPs and 20.64% fewer weight parameters on the SARS-CoV-2 CT-scan dataset. The results of our experiments showed that the proposed method significantly reduces the run-time resource requirements of the computationally intensive models and makes them ready to be utilized on the point-of-care devices.

Multi-task multi-scale learning for outcome prediction in 3D PET images

BACKGROUND AND OBJECTIVES

Predicting patient response to treatment and survival in oncology is a prominent way towards precision medicine. To this end, radiomics has been proposed as a field of study where images are used instead of invasive methods. The first step in radiomic analysis in oncology is lesion segmentation. However, this task is time consuming and can be physician subjective. Automated tools based on supervised deep learning have made great progress in helping physicians. However, they are data hungry, and annotated data remains a major issue in the medical field where only a small subset of annotated images are available.

METHODS

In this work, we propose a multi-task, multi-scale learning framework to predict patient's survival and response. We show that the encoder can leverage multiple tasks to extract meaningful and powerful features that improve radiomic performance. We also show that subsidiary tasks serve as an inductive bias so that the model can better generalize.

RESULTS

Our model was tested and validated for treatment response and survival in esophageal and lung cancers, with an area under the ROC curve of 77% and 71% respectively, outperforming single-task learning methods.

CONCLUSIONS

Multi-task multi-scale learning enables higher performance of radiomic analysis by extracting rich information from intratumoral and peritumoral regions.

Combined 18 F-FDG PET/CT Radiomics and Sarcopenia Score in Predicting Relapse-Free Survival and Overall Survival in Patients With Esophagogastric Cancer

PURPOSE

The aim of this study was to determine if radiomic features combined with sarcopenia measurements on pretreatment 18 F-FDG PET/CT can improve outcome prediction in surgically treated adenocarcinoma esophagogastric cancer patients.

PATIENTS AND METHODS

One hundred forty-five esophageal adenocarcinoma patients with curative therapeutic intent and available pretreatment 18 F-FDG PET/CT were included. Textural features from PET and CT images were evaluated using LIFEx software ( lifexsoft.org ). Sarcopenia measurements were done by measuring the Skeletal Muscle Index at L3 level on the CT component. Univariable and multivariable analyses were conducted to create a model including the radiomic parameters, clinical features, and Skeletal Muscle Index score to predict patients' outcome.

RESULTS

In multivariable analysis, we combined clinicopathological parameters including ECOG, surgical T, and N staging along with imaging derived sarcopenia measurements and radiomic features to build a predictor model for relapse-free survival and overall survival. Overall, adding sarcopenic status to the model with clinical features only (likelihood ratio test P = 0.03) and CT feature ( P = 0.0037) improved the model fit for overall survival. Similarly, adding sarcopenic status ( P = 0.051), CT feature ( P = 0.042), and PET feature ( P = 0.011) improved the model fit for relapse-free survival.

CONCLUSIONS

PET and CT radiomics derived from combined PET/CT integrated with clinicopathological parameters and sarcopenia measurement might improve outcome prediction in patients with nonmetastatic esophagogastric adenocarcinoma.

Domestic pig sound classification based on TransformerCNN

Excellent performance has been demonstrated in implementing challenging agricultural production processes using modern information technology, especially in the use of artificial intelligence methods to improve modern production environments. However, most of the existing work uses visual methods to train models that extract image features of organisms to analyze their behavior, and it may not be truly intelligent. Because vocal animals transmit information through grunts, the information obtained directly from the grunts of pigs is more useful to understand their behavior and emotional state, which is important for monitoring and predicting the health conditions and abnormal behavior of pigs. We propose a sound classification model called TransformerCNN, which combines the advantages of CNN spatial feature representation and the Transformer sequence coding to form a powerful global feature perception and local feature extraction capability. Through detailed qualitative and quantitative evaluations and by comparing state-of-the-art traditional animal sound recognition methods with deep learning methods, we demonstrate the advantages of our approach for classifying domestic pig sounds. The scores for domestic pig sound recognition accuracy, AUC and recall were 96.05%, 98.37% and 90.52%, respectively, all higher than the comparison model. In addition, it has good robustness and generalization capability with low variation in performance for different input features.

Urban Land Use and Land Cover Change Analysis Using Random Forest Classification of Landsat Time Series

Efficient implementation of remote sensing image classification can facilitate the extraction of spatiotemporal information for land use and land cover (LULC) classification. Mapping LULC change can pave the way to investigate the impacts of different socioeconomic and environmental factors on the Earth’s surface. This study presents an algorithm that uses Landsat time-series data to analyze LULC change. We applied the Random Forest (RF) classifier, a robust classification method, in the Google Earth Engine (GEE) using imagery from Landsat 5, 7, and 8 as inputs for the 1985 to 2019 period. We also explored the performance of the pan-sharpening algorithm on Landsat bands besides the impact of different image compositions to produce a high-quality LULC map. We used a statistical pan-sharpening algorithm to increase multispectral Landsat bands’ (Landsat 7–9) spatial resolution from 30 m to 15 m. In addition, we checked the impact of different image compositions based on several spectral indices and other auxiliary data such as digital elevation model (DEM) and land surface temperature (LST) on final classification accuracy based on several spectral indices and other auxiliary data on final classification accuracy. We compared the classification result of our proposed method and the Copernicus Global Land Cover Layers (CGLCL) map to verify the algorithm. The results show that: (1) Using pan-sharpened top-of-atmosphere (TOA) Landsat products can produce more accurate results for classification instead of using surface reflectance (SR) alone; (2) LST and DEM are essential features in classification, and using them can increase final accuracy; (3) the proposed algorithm produced higher accuracy (94.438% overall accuracy (OA), 0.93 for Kappa, and 0.93 for F1-score) than CGLCL map (84.4% OA, 0.79 for Kappa, and 0.50 for F1-score) in 2019; (4) the total agreement between the classification results and the test data exceeds 90% (93.37–97.6%), 0.9 (0.91–0.96), and 0.85 (0.86–0.95) for OA, Kappa values, and F1-score, respectively, which is acceptable in both overall and Kappa accuracy. Moreover, we provide a code repository that allows classifying Landsat 4, 5, 7, and 8 within GEE. This method can be quickly and easily applied to other regions of interest for LULC mapping.

Radiomics in Oncological PET Imaging: A Systematic Review-Part 2, Infradiaphragmatic Cancers, Blood Malignancies, Melanoma and Musculoskeletal Cancers

The objective of this review was to summarize published radiomics studies dealing with infradiaphragmatic cancers, blood malignancies, melanoma, and musculoskeletal cancers, and assess their quality. PubMed database was searched from January 1990 to February 2022 for articles performing radiomics on PET imaging of at least 1 specified tumor type. Exclusion criteria includd: non-oncological studies; supradiaphragmatic tumors; reviews, comments, cases reports; phantom or animal studies; technical articles without a clinically oriented question; studies including <30 patients in the training cohort. The review database contained PMID, first author, year of publication, cancer type, number of patients, study design, independent validation cohort and objective. This database was completed twice by the same person; discrepant results were resolved by a third reading of the articles. A total of 162 studies met inclusion criteria; 61 (37.7%) studies included >100 patients, 13 (8.0%) were prospective and 61 (37.7%) used an independent validation set. The most represented cancers were esophagus, lymphoma, and cervical cancer (n = 24, n = 24 and n = 19 articles, respectively). Most studies focused on 18F-FDG, and prognostic and response to treatment objectives. Although radiomics and artificial intelligence are technically challenging, new contributions and guidelines help improving research quality over the years and pave the way toward personalized medicine.

Cancer Detection and Prediction Using Genetic Algorithms

Cancer is a wide category of diseases that is caused by the abnormal, uncontrollable growth of cells, and it is the second leading cause of death globally. Screening, early diagnosis, and prediction of recurrence give patients the best possible chance for successful treatment. However, these tests can be expensive and invasive and the results have to be interpreted by experts. Genetic algorithms (GAs) are metaheuristics that belong to the class of evolutionary algorithms. GAs can find the optimal or near-optimal solutions in huge, difficult search spaces and are widely used for search and optimization. This makes them ideal for detecting cancer by creating models to interpret the results of tests, especially noninvasive. In this article, we have comprehensively reviewed the existing literature, analyzed them critically, provided a comparative analysis of the state-of-the-art techniques, and identified the future challenges in the development of such techniques by medical professionals.

Weakly Supervised Tumor Detection in PET Using Class Response for Treatment Outcome Prediction

It is proven that radiomic characteristics extracted from the tumor region are predictive. The first step in radiomic analysis is the segmentation of the lesion. However, this task is time consuming and requires a highly trained physician. This process could be automated using computer-aided detection (CAD) tools. Current state-of-the-art methods are trained in a supervised learning setting, which requires a lot of data that are usually not available in the medical imaging field. The challenge is to train one model to segment different types of tumors with only a weak segmentation ground truth. In this work, we propose a prediction framework including a 3D tumor segmentation in positron emission tomography (PET) images, based on a weakly supervised deep learning method, and an outcome prediction based on a 3D-CNN classifier applied to the segmented tumor regions. The key step is to locate the tumor in 3D. We propose to (1) calculate two maximum intensity projection (MIP) images from 3D PET images in two directions, (2) classify the MIP images into different types of cancers, (3) generate the class activation maps through a multitask learning approach with a weak prior knowledge, and (4) segment the 3D tumor region from the two 2D activation maps with a proposed new loss function for the multitask. The proposed approach achieves state-of-the-art prediction results with a small data set and with a weak segmentation ground truth. Our model was tested and validated for treatment response and survival in lung and esophageal cancers on 195 patients, with an area under the receiver operating characteristic curve (AUC) of 67% and 59%, respectively, and a dice coefficient of 73% and 0.77% for tumor segmentation.

Survival risk prediction model for ESCC based on relief feature selection and CNN

Esophageal squamous cell carcinoma (ESCC) is a common malignant tumor of the digestive system with poor prognosis and high mortality. It is of great significance to predict the prognosis risk of patients with cancer by using medical pathology information. To take full advantage of the clinic pathological information of ESCC patients and improve the accuracy of postoperative survival risk prediction, this paper proposes an ESCC survival risk prediction model based on Relief feature selection and convolutional neural network (CNN). Firstly, statistical analysis methods and relief feature selection algorithm are used to extract the important risk factors related to the survival risk of patients. Then, One-dimensional convolutional neural network (1D-CNN) is used to establish the survival risk prediction model of patients with esophageal cancer. Finally, the data of patients with esophageal cancer provided by the First Affiliated Hospital of Zhengzhou University is used to assess the performance of the model. The results show that the model proposed in this paper has a high accuracy rate, which can effectively predict the postoperative survival risk of the patient through the clinical phenotypic index of the patient.

Methodological quality of machine learning-based quantitative imaging analysis studies in esophageal cancer: a systematic review of clinical outcome prediction after concurrent chemoradiotherapy

PURPOSE

Studies based on machine learning-based quantitative imaging techniques have gained much interest in cancer research. The aim of this review is to critically appraise the existing machine learning-based quantitative imaging analysis studies predicting outcomes of esophageal cancer after concurrent chemoradiotherapy in accordance with PRISMA guidelines.

METHODS

A systematic review was conducted in accordance with PRISMA guidelines. The citation search was performed via PubMed and Embase Ovid databases for literature published before April 2021. From each full-text article, study characteristics and model information were summarized. We proposed an appraisal matrix with 13 items to assess the methodological quality of each study based on recommended best-practices pertaining to quality.

RESULTS

Out of 244 identified records, 37 studies met the inclusion criteria. Study endpoints included prognosis, treatment response, and toxicity after concurrent chemoradiotherapy with reported discrimination metrics in validation datasets between 0.6 and 0.9, with wide variation in quality. A total of 30 studies published within the last 5 years were evaluated for methodological quality and we found 11 studies with at least 6 "good" item ratings.

CONCLUSION

A substantial number of studies lacked prospective registration, external validation, model calibration, and support for use in clinic. To further improve the predictive power of machine learning-based models and translate into real clinical applications in cancer research, appropriate methodologies, prospective registration, and multi-institution validation are recommended.

A different overview of staging PET/CT images in patients with esophageal cancer: the role of textural analysis with machine learning methods

OBJECTIVE

This study evaluates the ability of several machine learning (ML) algorithms, developed using volumetric and texture data extracted from baseline 18F-FDG PET/CT studies performed initial staging of patient with esophageal cancer (EC), to predict survival and histopathology.

METHODS

The initial staging 18F-FDG PET/CT images obtained on newly diagnosed EC patients between January 2008 and June 2019 were evaluated using LIFEx software. A region of interest (ROI) of the primary tumor was created and volumetric and textural features were obtained. A significant relationship between these features and pathological subtypes, 1-year, and 5-year survival was investigated. Due to the nonhomogeneity of the data, nonparametric test (The Mann-Whitney U test) was used for each feature, in pairwise comparisons of independent variables. A p value of < 0.05 was considered significant. Receiver operating curve (ROC) analysis was performed for features with p < 0.05. Correlation between the significant features was evaluated with Spearman correlation test; features with correlation coefficient < 0.8 were evaluated with several ML algorithms.

RESULTS

In predicting survival in a 1-year follow-up J48 was obtained as the most successful algorithm (AUC: 0.581, PRC: 0.565, MCC: 0.258, acc: 64.29%). 5-year survival results were more promising than 1-year survival results with (AUC: 0.820, PRC: 0.860, MCC: 271, acc: 81.36%) by logistic regression. It is revealed that the most successful algorithm was naive bayes (AUC: 0.680 PRC: 0.776, MCC: 0.298, acc: 82.66%) in the histopathological discrimination.

CONCLUSION

Texture analysis with ML algorithms could be predictive of overall survival and discriminating histopathological subtypes of EC.

Current-Visit and Next-Visit Prediction for Fatty Liver Disease With a Large-Scale Dataset: Model Development and Performance Comparison

BACKGROUND

Fatty liver disease (FLD) arises from the accumulation of fat in the liver and may cause liver inflammation, which, if not well controlled, may develop into liver fibrosis, cirrhosis, or even hepatocellular carcinoma.

OBJECTIVE

We describe the construction of machine-learning models for current-visit prediction (CVP), which can help physicians obtain more information for accurate diagnosis, and next-visit prediction (NVP), which can help physicians provide potential high-risk patients with advice to effectively prevent FLD.

METHODS

The large-scale and high-dimensional dataset used in this study comes from Taipei MJ Health Research Foundation in Taiwan. We used one-pass ranking and sequential forward selection (SFS) for feature selection in FLD prediction. For CVP, we explored multiple models, including k-nearest-neighbor classifier (KNNC), Adaboost, support vector machine (SVM), logistic regression (LR), random forest (RF), Gaussian naïve Bayes (GNB), decision trees C4.5 (C4.5), and classification and regression trees (CART). For NVP, we used long short-term memory (LSTM) and several of its variants as sequence classifiers that use various input sets for prediction. Model performance was evaluated based on two criteria: the accuracy of the test set and the intersection over union/coverage between the features selected by one-pass ranking/SFS and by domain experts. The accuracy, precision, recall, F-measure, and area under the receiver operating characteristic curve were calculated for both CVP and NVP for males and females, respectively.

RESULTS

After data cleaning, the dataset included 34,856 and 31,394 unique visits respectively for males and females for the period 2009-2016. The test accuracy of CVP using KNNC, Adaboost, SVM, LR, RF, GNB, C4.5, and CART was respectively 84.28%, 83.84%, 82.22%, 82.21%, 76.03%, 75.78%, and 75.53%. The test accuracy of NVP using LSTM, bidirectional LSTM (biLSTM), Stack-LSTM, Stack-biLSTM, and Attention-LSTM was respectively 76.54%, 76.66%, 77.23%, 76.84%, and 77.31% for fixed-interval features, and was 79.29%, 79.12%, 79.32%, 79.29%, and 78.36%, respectively, for variable-interval features.

CONCLUSIONS

This study explored a large-scale FLD dataset with high dimensionality. We developed FLD prediction models for CVP and NVP. We also implemented efficient feature selection schemes for current- and next-visit prediction to compare the automatically selected features with expert-selected features. In particular, NVP emerged as more valuable from the viewpoint of preventive medicine. For NVP, we propose use of feature set 2 (with variable intervals), which is more compact and flexible. We have also tested several variants of LSTM in combination with two feature sets to identify the best match for male and female FLD prediction. More specifically, the best model for males was Stack-LSTM using feature set 2 (with 79.32% accuracy), whereas the best model for females was LSTM using feature set 1 (with 81.90% accuracy).

Nonlinear Random Forest Classification, a Copula-Based Approach

In this work, we use a copula-based approach to select the most important features for a random forest classification. Based on associated copulas between these features, we carry out this feature selection. We then embed the selected features to a random forest algorithm to classify a label-valued outcome. Our algorithm enables us to select the most relevant features when the features are not necessarily connected by a linear function; also, we can stop the classification when we reach the desired level of accuracy. We apply this method on a simulation study as well as a real dataset of COVID-19 and for a diabetes dataset.

Predicting the Local Response of Esophageal Squamous Cell Carcinoma to Neoadjuvant Chemoradiotherapy by Radiomics with a Machine Learning Method Using 18F-FDG PET Images

Background: This study aimed to propose a machine learning model to predict the local response of resectable locally advanced esophageal squamous cell carcinoma (LA-ESCC) treated by neoadjuvant chemoradiotherapy (NCRT) using pretreatment 18-fluorodeoxyglucose positron emission tomography (FDG PET) images. Methods: The local responses of 98 patients were categorized into two groups (complete response and noncomplete response). We performed a radiomics analysis using five segmentations created on FDG PET images, resulting in 4250 features per patient. To construct a machine learning model, we used the least absolute shrinkage and selection operator (LASSO) regression to extract radiomics features optimal for the prediction. Then, a prediction model was constructed by using a neural network classifier. The training model was evaluated with 5-fold cross-validation. Results: By the LASSO analysis of the training data, 22 radiomics features were extracted. In the testing data, the average accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve score of the five prediction models were 89.6%, 92.7%, 89.5%, and 0.95, respectively. Conclusions: The proposed machine learning model using radiomics showed promising predictive accuracy of the local response of LA-ESCC treated by NCRT.

Identification of SARS-CoV-2 origin: Using Ngrams, principal component analysis and Random Forest algorithm

COVID-19 is an infectious disease caused by the newly discovered SARS-CoV-2 virus. This virus causes a respiratory tract infection, symptoms include dry cough, fever, tiredness and in more severe cases, breathing difficulty. SARS-CoV-2 is an extremely contagious virus that is spreading rapidly all over the world and the scientific community is working tirelessly to find an effective treatment. This paper aims to determine the origin of this virus by comparing its nucleic acid sequence with all members of the coronaviridae family. This study uses a new approach based on the combination of three powerful techniques which are: Ngrams (For text categorization), Principal Component Analysis (For dimensionality reduction) and Random Forest algorithm (For supervised classification). The experimental results have shown that a large set of SARS-CoV-2 genomes, collected from different locations around the world, present significant similarities to those found in pangolins. This finding confirms some previous results obtained by other methods, which also suggest that pangolins should be considered as possible hosts in the emergence of the new coronavirus.

Early Cancer Detection from Genome-wide Cell-free DNA Fragmentation via Shuffled Frog Leaping Algorithm and Support Vector Machine

MOTIVATION

Early cancer detection is significant for the patient mortality rate reduction. Although machine learning has been widely employed in that context, there are still deficiencies. In this work, we studied different machine learning algorithms for early cancer detection and proposed an Adaptive Support Vector Machine (ASVM) method by synergizing Shuffled Frog Leaping Algorithm (SFLA) and Support Vector Machine (SVM) in this paper.

RESULTS

As ASVM regulates SVM for parameter adaption based on data characteristics, the experimental results demonstrated the robust generalization capability of ASVM on different datasets under different settings; for instance, ASVM can enhance the sensitivity by over 10% for early cancer detection compared with SVM. Besides, our proposed ASVM outperformed Grid Search + SVM and Random Search + SVM by significant margins in terms of the area under the ROC curve (AUC) (0.938 vs. 0.922 vs. 0.921).

AVAILABILITY

The proposed algorithm and dataset are available at https://github.com/ElaineLIU-920/ASVM-for-Early-Cancer-Detection.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Deep learning based automatic diagnosis of first-episode psychosis, bipolar disorder and healthy controls

Neuroimaging data driven machine learning based predictive modeling and pattern recognition has been attracted strongly attention in biomedical sciences. Machine learning based diagnosis techniques are widely applied in diagnosis of neurological diseases. However, machine learning techniques are difficult to effectively extract deep information in neuroimaging data, resulting in low classification accuracy of mental illnesses. To address this problem, we propose a deep learning based automatic diagnosis first-episode psychosis (FEP), bipolar disorder (BD) and healthy controls (HC) method. Specifically, we design a convolutional neural network (CNN) framework to automatically diagnosis based on structural magnetic functional imaging (sMRI). Our dataset consists of 89 FEP patients, 40 BD patients and 83 HC. A three-way classifier (FEP vs. BD vs. HC) and three binary classifiers (FEP vs. BD, FEP vs. HC, BD vs. HC) are trained based on their gray matter volume images. Experiment results show that the performance of CNN-based method outperforms the classic classifiers both in two and three categories classification task. Our research reveals that abnormal gray matter volume is one of the main characteristics for discriminating FEP, BD and HC.

A Systematic Review of PET Textural Analysis and Radiomics in Cancer

Background: Although many works have supported the utility of PET radiomics, several authors have raised concerns over the robustness and replicability of the results. This study aimed to perform a systematic review on the topic of PET radiomics and the used methodologies. Methods: PubMed was searched up to 15 October 2020. Original research articles based on human data specifying at least one tumor type and PET image were included, excluding those that apply only first-order statistics and those including fewer than 20 patients. Each publication, cancer type, objective and several methodological parameters (number of patients and features, validation approach, among other things) were extracted. Results: A total of 290 studies were included. Lung (28%) and head and neck (24%) were the most studied cancers. The most common objective was prognosis/treatment response (46%), followed by diagnosis/staging (21%), tumor characterization (18%) and technical evaluations (15%). The average number of patients included was 114 (median = 71; range 20–1419), and the average number of high-order features calculated per study was 31 (median = 26, range 1–286). Conclusions: PET radiomics is a promising field, but the number of patients in most publications is insufficient, and very few papers perform in-depth validations. The role of standardization initiatives will be crucial in the upcoming years.

Heart Rate Information-Based Machine Learning Prediction of Emotions Among Pregnant Women

In this study, the extent to which different emotions of pregnant women can be predicted based on heart rate-relevant information as indicators of autonomic nervous system functioning was explored using various machine learning algorithms. Nine heart rate-relevant autonomic system indicators, including the coefficient of variation R-R interval (CVRR), standard deviation of all NN intervals (SDNN), and square root of the mean squared differences of successive NN intervals (RMSSD), were measured using a heart rate monitor (MyBeat) and four different emotions including "happy," as a positive emotion and "anxiety," "sad," "frustrated," as negative emotions were self-recorded on a smartphone application, during 1 week starting from 23rd to 32nd weeks of pregnancy from 85 pregnant women. The k-nearest neighbor (k-NN), support vector machine (SVM), logistic regression (LR), random forest (RF), naïve bayes (NB), decision tree (DT), gradient boosting trees (GBT), stochastic gradient descent (SGD), extreme gradient boosting (XGBoost), and artificial neural network (ANN) machine learning methods were applied to predict the four different emotions based on the heart rate-relevant information. To predict four different emotions, RF also showed a modest area under the receiver operating characteristic curve (AUC-ROC) of 0.70. CVRR, RMSSD, SDNN, high frequency (HF), and low frequency (LF) mostly contributed to the predictions. GBT displayed the second highest AUC (0.69). Comprehensive analyses revealed the benefits of the prediction accuracy of the RF and GBT methods and were beneficial to establish models to predict emotions based on autonomic nervous system indicators. The results implicated SDNN, RMSSD, CVRR, LF, and HF as important parameters for the predictions.

Transcriptomic biomarkers for predicting response to neoadjuvant treatment in oesophageal cancer

Oesophageal cancer is a devastating disease with poor outcomes and is the sixth leading cause of cancer death worldwide. In the setting of resectable disease, there is clear evidence that neoadjuvant chemotherapy and chemoradiotherapy result in improved survival. Disappointingly, only 15%-30% of patients obtain a histopathological response to neoadjuvant therapy, often at the expense of significant toxicity. There are no predictive biomarkers in routine clinical use in this setting and the ability to stratify patients for treatment could dramatically improve outcomes. In this review, we aim to outline current progress in evaluating predictive transcriptomic biomarkers for neoadjuvant therapy in oesophageal cancer and discuss the challenges facing biomarker development in this setting. We place these issues in the wider context of recommendations for biomarker development and reporting. The majority of studies focus on messenger RNA (mRNA) and microRNA (miRNA) biomarkers. These studies report a range of different genes involved in a wide variety of pathways and biological processes, and this is explained to a large extent by the different platforms and analysis methods used. Many studies are also vastly underpowered so are not suitable for identifying a candidate biomarker. Multiple molecular subtypes of oesophageal cancer have been proposed, although little is known about how these relate to clinical outcomes. We anticipate that the accumulating wealth of genomic and transcriptomic data and clinical trial collaborations in the coming years will provide unique opportunities to stratify patients in this poor-prognosis disease and recommend that future biomarker development incorporates well-designed retrospective and prospective analyses.

Multi-task deep learning based CT imaging analysis for COVID-19 pneumonia: Classification and segmentation

This paper presents an automatic classification segmentation tool for helping screening COVID-19 pneumonia using chest CT imaging. The segmented lesions can help to assess the severity of pneumonia and follow-up the patients. In this work, we propose a new multitask deep learning model to jointly identify COVID-19 patient and segment COVID-19 lesion from chest CT images. Three learning tasks: segmentation, classification and reconstruction are jointly performed with different datasets. Our motivation is on the one hand to leverage useful information contained in multiple related tasks to improve both segmentation and classification performances, and on the other hand to deal with the problems of small data because each task can have a relatively small dataset. Our architecture is composed of a common encoder for disentangled feature representation with three tasks, and two decoders and a multi-layer perceptron for reconstruction, segmentation and classification respectively. The proposed model is evaluated and compared with other image segmentation techniques using a dataset of 1369 patients including 449 patients with COVID-19, 425 normal ones, 98 with lung cancer and 397 of different kinds of pathology. The obtained results show very encouraging performance of our method with a dice coefficient higher than 0.88 for the segmentation and an area under the ROC curve higher than 97% for the classification.

A descriptive study of random forest algorithm for predicting COVID-19 patients outcome

Background

The outbreak of coronavirus disease 2019 (COVID-19) that occurred in Wuhan, China, has become a global public health threat. It is necessary to identify indicators that can be used as optimal predictors for clinical outcomes of COVID-19 patients.

Methods

The clinical information from 126 patients diagnosed with COVID-19 were collected from Wuhan Fourth Hospital. Specific clinical characteristics, laboratory findings, treatments and clinical outcomes were analyzed from patients hospitalized for treatment from 1 February to 15 March 2020, and subsequently died or were discharged. A random forest (RF) algorithm was used to predict the prognoses of COVID-19 patients and identify the optimal diagnostic predictors for patients' clinical prognoses.

Results

Seven of the 126 patients were excluded for losing endpoints, 103 of the remaining 119 patients were discharged (alive) and 16 died in the hospital. A synthetic minority over-sampling technique (SMOTE) was used to correct the imbalanced distribution of clinical patients. Recursive feature elimination (RFE) was used to select the optimal subset for analysis. Eleven clinical parameters, Myo, CD8, age, LDH, LMR, CD45, Th/Ts, dyspnea, NLR, D-Dimer and CK were chosen with AUC approximately 0.9905. The RF algorithm was built to predict the prognoses of COVID-19 patients based on the best subset, and the area under the ROC curve (AUC) of the test data was 100%. Moreover, two optimal clinical risk predictors, lactate dehydrogenase (LDH) and Myoglobin (Myo), were selected based on the Gini index. The univariable logistic analysis revealed a substantial increase in the risk for in-hospital mortality when Myo was higher than 80 ng/ml (OR = 7.54, 95% CI [3.42-16.63]) and LDH was higher than 500 U/L (OR = 4.90, 95% CI [2.13-11.25]).

Conclusion

We applied an RF algorithm to predict the mortality of COVID-19 patients with high accuracy and identified LDH higher than 500 U/L and Myo higher than 80 ng/ml to be potential risk factors for the prognoses of COVID-19 patients in the early stage of the disease.

Role of precision imaging in esophageal cancer

Esophageal cancer is a major cause of morbidity and mortality worldwide. Recent advancements in the management of esophageal cancer have allowed for earlier detection, improved ability to monitor progression, and superior treatment options. These innovations allow treatment teams to formulate more customized management plans and have led to an increase in patient survival rates. For example, in order for the most effective management plan to be constructed, accurate staging must be performed to determine tumor resectability. This article reviews the multimodality imaging approach involved in making a diagnosis, staging, evaluating treatment response and detecting recurrence in esophageal cancer.

Ischemic stroke lesion detection, characterization and classification in CT images with optimal features selection

Ischemic stroke is the dominant disorder for mortality and morbidity. For immediate diagnosis and treatment plan of ischemic stroke, computed tomography (CT) images are used. This paper proposes a histogram bin based novel algorithm to segment the ischemic stroke lesion using CT and optimal feature group selection to classify normal and abnormal regions. Steps followed are pre-processing, segmentation, extracting texture features, feature ranking, feature grouping, classification and optimal feature group (FG) selection. The first order features, gray level run length matrix features, gray level co-occurrence matrix features and Hu's moment features are extracted. Classification is done using logistic regression (LR), support vector machine classifier (SVMC), random forest classifier (RFC) and neural network classifier (NNC). This proposed approach effectively detects ischemic stroke lesion with a classification accuracy of 88.77%, 97.86%, 99.79% and 99.79% obtained by the LR, SVMC, RFC and NNC when FG12 is opted, which is validated by fourfold cross validation.

A transcriptomic study for identifying cardia- and non-cardia-specific gastric cancer prognostic factors using genetic algorithm-based methods

Gastric cancer (GC) is a heterogeneous tumour with numerous differences of epidemiologic and clinicopathologic features between cardia cancer and non-cardia cancer. However, few studies were performed to construct site-specific GC prognostic models. In this study, we identified site-specific GC transcriptomic prognostic biomarkers using genetic algorithm (GA)-based support vector machine (GA-SVM) and GA-based Cox regression method (GA-Cox) in the Cancer Genome Atlas (TCGA) database. The area under time-dependent receive operating characteristic (ROC) curve (AUC) regarding 5-year survival and concordance index (C-index) was used to evaluate the predictive ability of Cox regression models. Finally, we identified 10 and 13 prognostic biomarkers for cardia cancer and non-cardia cancer, respectively. Compared to traditional models, the addition of these site-specific biomarkers could notably improve the model preference (cardia: AUCtraditional vs AUCcombined  = 0.720 vs 0.899, P = 8.75E-08; non-cardia: AUCtraditional vs AUCcombined  = 0.798 vs 0.994, P = 7.11E-16). The combined nomograms exhibited superior performance in cardia and non-cardia GC survival prediction (C-indexcardia  = 0.816; C-indexnoncardia  = 0.812). We also constructed a user-friendly GC site-specific molecular system (GC-SMS, https://njmu-zhanglab.shinyapps.io/gc_sms/), which is freely available for users. In conclusion, we developed site-specific GC prognostic models for predicting cardia cancer and non-cardia cancer survival, providing more support for the individualized therapy of GC patients.

CT radiomic features for predicting resectability of oesophageal squamous cell carcinoma as given by feature analysis: a case control study

Background

Computed tomography (CT) is commonly used in all stages of oesophageal squamous cell carcinoma (SCC) management. Compared to basic CT features, CT radiomic features can objectively obtain more information about intratumour heterogeneity. Although CT radiomics has been proved useful for predicting treatment response to chemoradiotherapy in oesophageal cancer, the best way to use CT radiomic biomarkers as predictive markers for determining resectability of oesophageal SCC remains to be developed. This study aimed to develop CT radiomic features related to resectability of oesophageal SCC with five predictive models and to determine the most predictive model.

Methods

Five hundred ninety-one patients with oesophageal SCC undergoing contrast-enhanced CT were enrolled in this study, and were composed by 270 resectable cases and 321 unresectable cases. Of the 270 resectable oesophageal SCCs, 91 cases were primary resectable tumours; and the remained 179 cases received neoadjuvant therapy after CT, shrank on therapy, and changed to resectable tumours. Four hundred thirteen oesophageal SCCs including 189 resectable cancers and 224 unresectable cancers were randomly allocated to the training cohort; and 178 oesophageal SCCs including 81 resectable tumours and 97 unresectable tumours were allocated to the validation group. Four hundred ninety-five radiomic features were extracted from CT data for identifying resectability of oesophageal SCC. Useful radiomic features were generated by dimension reduction using least absolute shrinkage and selection operator. The optimal radiomic features were chosen using multivariable logistic regression, random forest, support vector machine, X-Gradient boost and decision tree classifiers. Discriminating performance was assessed with area under receiver operating characteristic curve (AUC), accuracy and F-1score.

Results

Eight radiomic features were selected to create radiomic models related to resectability of oesophageal SCC (P-values < 0.01 for both cohorts). Multivariable logistic regression model showed the best performance (AUC = 0.92 ± 0.04 and 0.87 ± 0.02, accuracy = 0.87 and 0.86, and F-1score = 0.93 and 0.86 in training and validation cohorts, respectively) in comparison with any other model (P-value < 0.001). Good calibration was observed for multivariable logistic regression model.

Conclusion

CT radiomic models could help predict resectability of oesophageal SCC, and multivariable logistic regression model is the most predictive model.

A comparative study on feature selection for a risk prediction model for colorectal cancer

BACKGROUND AND OBJECTIVE

Risk prediction models aim at identifying people at higher risk of developing a target disease. Feature selection is particularly important to improve the prediction model performance avoiding overfitting and to identify the leading cancer risk (and protective) factors. Assessing the stability of feature selection/ranking algorithms becomes an important issue when the aim is to analyze the features with more prediction power.

METHODS

This work is focused on colorectal cancer, assessing several feature ranking algorithms in terms of performance for a set of risk prediction models (Neural Networks, Support Vector Machines (SVM), Logistic Regression, k-Nearest Neighbors and Boosted Trees). Additionally, their robustness is evaluated following a conventional approach with scalar stability metrics and a visual approach proposed in this work to study both similarity among feature ranking techniques as well as their individual stability. A comparative analysis is carried out between the most relevant features found out in this study and features provided by the experts according to the state-of-the-art knowledge.

RESULTS

The two best performance results in terms of Area Under the ROC Curve (AUC) are achieved with a SVM classifier using the top-41 features selected by the SVM wrapper approach (AUC=0.693) and Logistic Regression with the top-40 features selected by the Pearson (AUC=0.689). Experiments showed that performing feature selection contributes to classification performance with a 3.9% and 1.9% improvement in AUC for the SVM and Logistic Regression classifier, respectively, with respect to the results using the full feature set. The visual approach proposed in this work allows to see that the Neural Network-based wrapper ranking is the most unstable while the Random Forest is the most stable.

CONCLUSIONS

This study demonstrates that stability and model performance should be studied jointly as Random Forest turned out to be the most stable algorithm but outperformed by others in terms of model performance while SVM wrapper and the Pearson correlation coefficient are moderately stable while achieving good model performance.