Development, comparison, and internal validation of prediction models to determine the visual prognosis of patients with open globe injuries using machine learning approaches.

INTRODUCTION: Open globe injuries (OGI) represent a main preventable reason for blindness and visual impairment, particularly in developing countries. The goal of this study is evaluating key variables affecting the prognosis of open globe injuries and validating internally and comparing different machine learning models to estimate final visual acuity. MATERIALS AND METHODS: We reviewed three hundred patients with open globe injuries receiving treatment at Khatam-Al-Anbia Hospital in Iran from 2020 to 2022. Age, sex, type of trauma, initial VA grade, relative afferent pupillary defect (RAPD), zone of trauma, traumatic cataract, traumatic optic neuropathy (TON), intraocular foreign body (IOFB), retinal detachment (RD), endophthalmitis, and ocular trauma score (OTS) grade were the input features. We calculated univariate and multivariate regression models to assess the association of different features with visual acuity (VA) outcomes. We predicted visual acuity using ten supervised machine learning algorithms including multinomial logistic regression (MLR), support vector machines (SVM), K-nearest neighbors (KNN), naïve bayes (NB), decision tree (DT), random forest (RF), bagging (BG), adaptive boosting (ADA), artificial neural networks (ANN), and extreme gradient boosting (XGB). Accuracy, positive predictive value (PPV), recall, F-score, brier score (BS), Matthew correlation coefficient (MCC), receiver operating characteristic (AUC-ROC), and calibration plot were used to assess how well machine learning algorithms performed in predicting the final VA. RESULTS: The artificial neural network (ANN) model had the best accuracy to predict the final VA. The sensitivity, F1 score, PPV, accuracy, and MCC of the ANN model were 0.81, 0.85, 0.89, 0.93, and 0.81, respectively. In addition, the estimated AUC-ROC and AUR-PRC of the ANN model for OGI patients were 0.96 and 0.91, respectively. The brier score and calibration log-loss for the ANN model was 0.201 and 0.232, respectively. CONCLUSION: As classic and ensemble ML models were compared, results shows that the ANN model was the best. As a result, the framework that has been presented may be regarded as a good substitute for predicting the final VA in OGI patients. Excellent predictive accuracy was shown by the open globe injury model developed in this study, which should be helpful to provide clinical advice to patients and making clinical decisions concerning the management of open globe injuries.

Effect of a gamified mobile-based self-management application on disease activity index, quality of life, and mental health in adults with inflammatory bowel disease: A protocol of a randomized controlled trial study.

Background and Aims: Inflammatory bowel disease (IBD) is a chronic inflammatory gastrointestinal tract disease subdivided into Crohn's disease (CD) and ulcerative colitis (UC). There is currently no cure for IBD, and individuals with IBD frequently experience a lower health-related quality of life (HRQOL) than the general population. Gamification has become an increasingly popular topic in recent years. Adapting game design concepts to nongaming contexts represents a novel and potential approach to changing user engagement. This study will be conducted with the aim of evaluating the effect of a gamified mobile-based self-management application on disease activity index, quality of life, and mental health in adults with IBD. Methods: A multicenter, parallel, two-arm, exploratory randomized controlled trial with a 6-month follow-up per patient will be designed to compare the impact of the gamified mobile-based tele-management system on primary and secondary health outcomes and outpatient visits in 210 patients with all types of IBD which are divided equally into a control group with standard care and an intervention group which will use the developed mobile application named MY IBD BUDDY. All patients will attend study visits at baseline, 12 and 24 weeks, and routine IBD clinic visits or telephone consultations based on randomization group assignment. Disease activity or disease activity index, mental health (anxiety and depression) symptoms, quality of life, self-efficacy, and IBD-specific knowledge will be measured at baseline with two follow-ups at 12 and 24 weeks. Conclusions: In sum, the outcomes of our trial will demonstrate the impact of the gamified mobile-based self-management system on disease activity, quality of life, and anxiety and depression by means of interactive care and patient empowerment. Trial Registration: IRCT: IRCT20200613047757N1. Registered November 16, 2021. Prospectively registered and visible at OSF (https://doi.org/10.17605/OSF.IO/AWFY9).

Inflammatory bowel disease patients' perspectives of non-medical needs.

BACKGROUND: Inflammatory bowel disease (IBD) imposes a huge burden on the healthcare systems and greatly declines the patient's quality of life. However, there is a paucity of detailed data regarding information and supportive needs as well as sources and methods of obtaining information to control different aspects of the disease from the perspectives of the patients themselves. This study aimed to establish the IBD patients' preferences of informational and supportive needs through Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA). METHODS: IBD patients were recruited from different centers. Considering inclusion and exclusion criteria, 521 participants were filled a predefined questionnaire. This questionnaire was prepared through literature review of the recent well-known guidelines on the needs of IBD patients, which was further approved by the experts of IBD area in three rounds of Delphi consensus. It includes 56 items in four sections of informational needs (25), supportive needs (15), sources of information (7), and methods of obtaining information (9). RESULTS: In particular, EFA was used to apply data reduction and structure detection. Given that this study tries to identify patterns, structures as well as inter-relationships and classification of the variables, EFA was utilized to simplify presentation of the variables in a way that large amounts of observations transform into fewer ones. Accordingly, the EFA identified five factors out of 25 items in the information needs section, three factors out of 15 items in the supportive needs section, two factors out of 7 items in the information sources section, and two factors out of 9 items in the information presentation methods. Through the CFA, all 4 models were supported by Root Mean Squared Error of Approximation (RMSEA); Incremental Fit Index (IFI); Comparative Fit Index (CFI); Tucker-Lewis Index (TLI); and SRMR. These values were within acceptable ranges, indicating that the twelve factors achieved from EFA were validated. CONCLUSIONS: This study introduced a reliable 12-factor model as an efficient tool to comprehensively identify preferences of IBD patients in informational and supportive needs along with sources and methods of obtaining information. An in-depth understanding of the needs of IBD patients facilitates informing and supporting health service provision. It also assists patients in a fundamental way to improve adaptation and increase the quality of life. We suggest that health care providers consider the use of this tool in clinical settings in order to precisely assess its efficacy.

Models to predict length of stay in the emergency department: a systematic literature review and appraisal.

INTRODUCTION: Prolonged Length of Stay (LOS) in ED (Emergency Department) has been associated with poor clinical outcomes. Prediction of ED LOS may help optimize resource utilization, clinical management, and benchmarking. This study aims to systematically review models for predicting ED LOS and to assess the reporting and methodological quality about these models. METHODS: The online database PubMed, Scopus, and Web of Science (10 Sep 2023) was searched for English language articles that reported prediction models of LOS in ED. Identified titles and abstracts were independently screened by two reviewers. All original papers describing either development (with or without internal validation) or external validation of a prediction model for LOS in ED were included. RESULTS: Of 12,193 uniquely identified articles, 34 studies were included (29 describe the development of new models and five describe the validation of existing models). Different statistical and machine learning methods were applied to the papers. On the 39-point reporting score and 11-point methodological quality score, the highest reporting scores for development and validation studies were 39 and 8, respectively. CONCLUSION: Various studies on prediction models for ED LOS were published but they are fairly heterogeneous and suffer from methodological and reporting issues. Model development studies were associated with a poor to a fair level of methodological quality in terms of the predictor selection approach, the sample size, reproducibility of the results, missing imputation technique, and avoiding dichotomizing continuous variables. Moreover, it is recommended that future investigators use the confirmed checklist to improve the quality of reporting.

A systematic review of machine learning-based tumor-infiltrating lymphocytes analysis in colorectal cancer: Overview of techniques, performance metrics, and clinical outcomes.

The incidence of colorectal cancer (CRC), one of the deadliest cancers around the world, is increasing. Tissue microenvironment (TME) features such as tumor-infiltrating lymphocytes (TILs) can have a crucial impact on diagnosis or decision-making for treating patients with CRC. While clinical studies showed that TILs improve the host immune response, leading to a better prognosis, inter-observer agreement for quantifying TILs is not perfect. Incorporating machine learning (ML) based applications in clinical routine may promote diagnosis reliability. Recently, ML has shown potential for making progress in routine clinical procedures. We aim to systematically review the TILs analysis based on ML in CRC histological images. Deep learning (DL) and non-DL techniques can aid pathologists in identifying TILs, and automated TILs are associated with patient outcomes. However, a large multi-institutional CRC dataset with a diverse and multi-ethnic population is necessary to generalize ML methods.

A comparative study of explainable ensemble learning and logistic regression for predicting in-hospital mortality in the emergency department.

This study addresses the challenges associated with emergency department (ED) overcrowding and emphasizes the need for efficient risk stratification tools to identify high-risk patients for early intervention. While several scoring systems, often based on logistic regression (LR) models, have been proposed to indicate patient illness severity, this study aims to compare the predictive performance of ensemble learning (EL) models with LR for in-hospital mortality in the ED. A cross-sectional single-center study was conducted at the ED of Imam Reza Hospital in northeast Iran from March 2016 to March 2017. The study included adult patients with one to three levels of emergency severity index. EL models using Bagging, AdaBoost, random forests (RF), Stacking and extreme gradient boosting (XGB) algorithms, along with an LR model, were constructed. The training and validation visits from the ED were randomly divided into 80% and 20%, respectively. After training the proposed models using tenfold cross-validation, their predictive performance was evaluated. Model performance was compared using the Brier score (BS), The area under the receiver operating characteristics curve (AUROC), The area and precision-recall curve (AUCPR), Hosmer-Lemeshow (H-L) goodness-of-fit test, precision, sensitivity, accuracy, F1-score, and Matthews correlation coefficient (MCC). The study included 2025 unique patients admitted to the hospital's ED, with a total percentage of hospital deaths at approximately 19%. In the training group and the validation group, 274 of 1476 (18.6%) and 152 of 728 (20.8%) patients died during hospitalization, respectively. According to the evaluation of the presented framework, EL models, particularly Bagging, predicted in-hospital mortality with the highest AUROC (0.839, CI (0.802-0.875)) and AUCPR = 0.64 comparable in terms of discrimination power with LR (AUROC (0.826, CI (0.787-0.864)) and AUCPR = 0.61). XGB achieved the highest precision (0.83), sensitivity (0.831), accuracy (0.842), F1-score (0.833), and the highest MCC (0.48). Additionally, the most accurate models in the unbalanced dataset belonged to RF with the lowest BS (0.128). Although all studied models overestimate mortality risk and have insufficient calibration (P > 0.05), stacking demonstrated relatively good agreement between predicted and actual mortality. EL models are not superior to LR in predicting in-hospital mortality in the ED. Both EL and LR models can be considered as screening tools to identify patients at risk of mortality.

Author Response.

How to cite this article: Rahmatinejad Z, Hoseini B, Pourmand A, Reihani H, Rahmatinejad F, Eslami S, et al. Author Response. Indian J Crit Care Med 2024;28(2):183-184.

The effect of computerised decision support alerts tailored to intensive care on the administration of high-risk drug combinations, and their monitoring: a cluster randomised stepped-wedge trial.

BACKGROUND: Drug-drug interactions (DDIs) can harm patients admitted to the intensive care unit (ICU). Yet, clinical decision support systems (CDSSs) aimed at helping physicians prevent DDIs are plagued by low-yield alerts, causing alert fatigue and compromising patient safety. The aim of this multicentre study was to evaluate the effect of tailoring potential DDI alerts to the ICU setting on the frequency of administered high-risk drug combinations. METHODS: We implemented a cluster randomised stepped-wedge trial in nine ICUs in the Netherlands. Five ICUs already used potential DDI alerts. Patients aged 18 years or older admitted to the ICU with at least two drugs administered were included. Our intervention was an adapted CDSS, only providing alerts for potential DDIs considered as high risk. The intervention was delivered at the ICU level and targeted physicians. We hypothesised that showing only relevant alerts would improve CDSS effectiveness and lead to a decreased number of administered high-risk drug combinations. The order in which the intervention was implemented in the ICUs was randomised by an independent researcher. The primary outcome was the number of administered high-risk drug combinations per 1000 drug administrations per patient and was assessed in all included patients. This trial was registered in the Netherlands Trial Register (identifier NL6762) on Nov 26, 2018, and is now closed. FINDINGS: In total, 10 423 patients admitted to the ICU between Sept 1, 2018, and Sept 1, 2019, were assessed and 9887 patients were included. The mean number of administered high-risk drug combinations per 1000 drug administrations per patient was 26·2 (SD 53·4) in the intervention group (n=5534), compared with 35·6 (65·0) in the control group (n=4353). Tailoring potential DDI alerts to the ICU led to a 12% decrease (95% CI 5-18%; p=0·0008) in the number of administered high-risk drug combinations per 1000 drug administrations per patient, after adjusting for clustering and prognostic factors. INTERPRETATION: This cluster randomised stepped-wedge trial showed that tailoring potential DDI alerts to the ICU setting significantly reduced the number of administered high-risk drug combinations. Our list of high-risk drug combinations can be used in other ICUs, and our strategy of tailoring alerts based on clinical relevance could be applied to other clinical settings. FUNDING: ZonMw.

Development and Validation of an Educational Book on Self-Management in Inflammatory Bowel Disease Based on Patient Preferences and Expert Opinions: A Methodological Study.

BACKGROUND: Self-management education resources for inflammatory bowel disease (IBD) using concepts remain infrequent. We aim to describe the development and evaluation process of educational material for self-management in IBD based on patient preferences and expert opinions. RESEARCH DESIGN AND METHODS: The method of this study includes two main phases of development and validation in five steps in the following order: (1) identification of information needs for patients with IBD; (2) content development with a comprehensive literature review and scientific texts related to IBD; (3) measuring the face validity of the content based on the expert opinions in the field of IBD; (4) validation of the content with the experts in the field of IBD; and (5) validation by target audiences. RESULTS: The expert panel comprises ten gastroenterologists, nutritionists, psychologists, gynecologists, and nurses. The total suitability score is 79.5%. The final draft version of the educational self-management material was presented to 30 IBD patients who were satisfied (n = 24; 80%) with the material. CONCLUSIONS: This study shows the development process and is validated for face and content validity by the academic multidisciplinary expert panel and target group. Patients and their caregivers can use this content to cope with their disease.

Computer-aided diagnosis of keratoconus through VAE-augmented images using deep learning.

Detecting clinical keratoconus (KCN) poses a challenging and time-consuming task. During the diagnostic process, ophthalmologists are required to review demographic and clinical ophthalmic examinations in order to make an accurate diagnosis. This study aims to develop and evaluate the accuracy of deep convolutional neural network (CNN) models for the detection of keratoconus (KCN) using corneal topographic maps. We retrospectively collected 1758 corneal images (978 normal and 780 keratoconus) from 1010 subjects of the KCN group with clinically evident keratoconus and the normal group with regular astigmatism. To expand the dataset, we developed a model using Variational Auto Encoder (VAE) to generate and augment images, resulting in a dataset of 4000 samples. Four deep learning models were used to extract and identify deep corneal features of original and synthesized images. We demonstrated that the utilization of synthesized images during training process increased classification performance. The overall average accuracy of the deep learning models ranged from 99% for VGG16 to 95% for EfficientNet-B0. All CNN models exhibited sensitivity and specificity above 0.94, with the VGG16 model achieving an AUC of 0.99. The customized CNN model achieved satisfactory results with an accuracy and AUC of 0.97 at a much faster processing speed compared to other models. In conclusion, the DL models showed high accuracy in screening for keratoconus based on corneal topography images. This is a development toward the potential clinical implementation of a more enhanced computer-aided diagnosis (CAD) system for KCN detection, which would aid ophthalmologists in validating the clinical decision and carrying out prompt and precise KCN treatment.

Expert opinions on informational and supportive needs and sources of obtaining information in patients with inflammatory bowel disease: a Delphi consensus study.

Background: The present study introduces informational and supportive needs and sources of obtaining information in patients with inflammatory bowel disease (IBD) through a three-round Expert Delphi Consensus Opinions method. Methods: According to our previous scoping review, important items in the area of informational and supportive needs and sources of obtaining information were elucidated. After omitting duplicates, 56 items in informational needs, 36 items in supportive needs, and 36 items in sources of obtaining information were retrieved. Both open- and close-ended questions were designed for each category in the form of three questionnaires. The questionnaires were sent to selected experts from different specialties. Experts responded to the questions in the first round. Based on the feedback, questions were modified and sent back to the experts in the second round. This procedure was repeated up to the third round. Results: In the first round, five items from informational needs, one item from supportive needs, and seven items from sources of obtaining information were identified as unimportant and omitted. Moreover, two extra items were proposed by the experts, which were added to the informational needs category. In the second round, seven, three, and seven items from informational needs, supportive needs, and sources of obtaining information were omitted due to the items being unimportant. In the third round, all the included items gained scores equal to or greater than the average and were identified as important. Kendall coordination coefficient W was calculated to be 0.344 for information needs, 0.330 for supportive needs, and 0.325 for sources of obtaining information, indicating a fair level of agreement between experts. Conclusions: Out of 128 items in the first round, the omission of 30 items and the addition of two items generated a 100-item questionnaire for three sections of informational needs, supportive needs, and sources of obtaining information with a high level of convergence between experts' viewpoints.

Application of ensemble machine learning approach to assess the factors affecting size and polydispersity index of liposomal nanoparticles.

Liposome nanoparticles have emerged as promising drug delivery systems due to their unique properties. Assessing particle size and polydispersity index (PDI) is critical for evaluating the quality of these liposomal nanoparticles. However, optimizing these parameters in a laboratory setting is both costly and time-consuming. This study aimed to apply a machine learning technique to assess the impact of specific factors, including sonication time, extrusion temperature, and compositions, on the size and PDI of liposomal nanoparticles. Liposomal solutions were prepared and subjected to sonication with varying values for these parameters. Two compositions: (A) HSPC:DPPG:Chol:DSPE-mPEG2000 at 55:5:35:5 molar ratio and (B) HSPC:Chol:DSPE-mPEG2000 at 55:40:5 molar ratio, were made using remote loading method. Ensemble learning (EL), a machine learning technique, was employed using the Least-squares boosting (LSBoost) algorithm to accurately model the data. The dataset was randomly split into training and testing sets, with 70% allocated for training. The LSBoost algorithm achieved mean absolute errors of 1.652 and 0.0105 for modeling the size and PDI, respectively. Under conditions where the temperature was set at approximately 60 °C, our EL model predicted a minimum particle size of 116.53 nm for composition (A) with a sonication time of approximately 30 min. Similarly, for composition (B), the model predicted a minimum particle size of 129.97 nm with sonication times of approximately 30 or 55 min. In most instances, a PDI of less than 0.2 was achieved. These results highlight the significant impact of optimizing independent factors on the characteristics of liposomal nanoparticles and demonstrate the potential of EL as a decision support system for identifying the best liposomal formulation. We recommend further studies to explore the effects of other independent factors, such as lipid composition and surfactants, on liposomal nanoparticle characteristics.

The effects of low-fat dairy products fortified with 1500 IU vitamin D3 on serum liver function biomarkers in adults with abdominal obesity: a randomized controlled trial.

INTRODUCTION: Vitamin D deficiency has been reported to affect liver function biomarkers. This study was aimed to investigate the effect of consuming vitamin D fortified low-fat dairy products on liver function tests in adults with abdominal obesity. METHODS: This total blinded randomized controlled trial was undertaken on otherwise healthy abdominally obese adults living in Mashhad, Iran. Milk and yogurt were fortified with 1500 IU vitamin D3 nano-capsules. Participants were randomized to receive fortified milk (n = 73), plain milk (n = 73), fortified yogurt (n = 69), and plain yogurt (n = 74) for 10 weeks. Blood samples were taken at baseline and at the end of the study to assess serum levels of vitamin D, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase (ALP), and Gamma glutamyl transferase. RESULTS: A total of 289 participants completed the study (54% female). The groups were homogenous in terms of age, sex, weight, energy intake, and physical activity level (p-value > 0.05). After the trial, vitamin D serum levels were significantly increased in both groups receiving fortified products (both p < 0.001). There was a significant time*group effect only in serum ALP (p < 0.001). CONCLUSION: Consumption of dairy products fortified by 1500 IU vitamin D3 might have detrimental effects on serum levels of some liver enzymes in individuals with abdominal obesity. Further studies needed to determine these effects and underlying mechanisms. TRIAL REGISTRATION: IRCT20101130005280N27 .

Optimizing nanoliposomal formulations: Assessing factors affecting entrapment efficiency of curcumin-loaded liposomes using machine learning.

BACKGROUND: Curcumin faces challenges in clinical applications due to its low bioavailability and poor water solubility. Liposomes have emerged as a promising delivery system for curcumin. This study aims to apply ensemble learning, a machine learning technique, to determine the most effective experimental conditions for formulating stable curcumin-loaded liposomes with a high entrapment efficiency (EE). METHODS: Two liposomal formulations composed of HSPC:DPPG:Chol:DSPE-mPEG2000 and HSPC:Chol:DSPE-mPEG2000 at 55:5:35:5 and 55:40:5 M ratios, respectively, were prepared using the remote loading method, and their particle size and polydispersity index (PDI) were determined using Dynamic Light Scattering. To model the impact of five factors (molar ratios, particle size, sonication time, pH, and PDI) on EE%, the Least-squares boosting (LSBoost) ensemble learning algorithm was employed due to its capability to effectively handle nonlinear and non-stationary problems. The implementation and optimization of LSBoost were performed using MATLAB R2020a. The dataset was randomly split into training and testing sets, with 70% allocated for training. The mean absolute error (MAE) was used as the cost function to evaluate model performance. Additionally, a novel approach was employed to visualize the results using 3D plots, facilitating practical interpretation. RESULTS: The optimal model exhibited an MAE of 3.61, indicating its robust predictive capability. The study identified several optimal conditions for achieving the highest EE value of 100%. However, to ensure both the highest EE value and a suitable particle size, it is recommended to set the following conditions: a molar ratio of 55:5:35:5, a PDI within the range of 0.09-0.13, a particle size of approximately 130 nm, a sonication time of 30 min, and a pH within the range of 7.2-8. It is worth mentioning that adjusting the molar ratio to 55:40:5 resulted in a maximum EE of 88.38%. CONCLUSION: These findings underscore the high performance of ensemble learning in accurately predicting and optimizing the EE of the curcumin-loaded liposomes. The application of this technique provides valuable insights and holds promise for the development of efficient drug delivery systems.

Association of time-restricted feeding, arterial age, and arterial stiffness in adults with metabolic syndrome.

Background: Time-restricted feeding (TRF) is a kind of intermittent fasting defined as eating and drinking only during a certain number of hours in a day. It has been suggested that intermittent fasting may improve cardiovascular risk factors. This study evaluated the association of TRF and arterial stiffness, using pulse wave velocity (PWV), pulse wave analysis, and arterial age in metabolic syndrome participants. Methods: A cohort study was carried out among metabolic syndrome adults who were followed over the Ramadan fasting period (used as a model of TRF since food was only allowed for about 8 h/day). The subjects were divided into Ramadan fasting and Ramadan nonfasting groups. The aortic PWV and central aortic pressure waveform were measured. Central systolic pressure, central pulse pressure, and indices of arterial compliance, such as augmentation pressure and augmentation index (AIx), were determined from waveform analysis. Results: Ninety-five adults (31.57% female, age: 45.46 ± 9.10 years) with metabolic syndrome (based on the International Diabetes Federation definition) participated in this study. Ramadan fasting and Ramadan nonfasting groups were including 80 and 15 individuals respectively. A significant reduction was seen in PWV (0.29 m/s), central systolic pressure (4.03 mmHg), central pulse pressure (2.43 mmHg), central augmentation pressure (1.88 mmHg), and central AIx (2.47) in the Ramadan fasting group (p = 0.014, p < 0.001, p = 0.001, p = 0.003, and p = 0.036 respectively). There were no significant changes in these indices among the Ramadan nonfasting group. Conclusions: This study suggested that TRF reduces arterial age and improves arterial stiffness among people with metabolic syndrome. This might be considered a beneficial nutrition strategy for extending healthspan (and perhaps longevity).

Assessment of the Tumor-Stroma Ratio and Tumor-Infiltrating Lymphocytes in Colorectal Cancer: Inter-Observer Agreement Evaluation.

BACKGROUND: To implement the new marker in clinical practice, reliability assessment, validation, and standardization of utilization must be applied. This study evaluated the reliability of tumor-infiltrating lymphocytes (TILs) and tumor-stroma ratio (TSR) assessment through conventional microscopy by comparing observers' estimations. METHODS: Intratumoral and tumor-front stromal TILs, and TSR, were assessed by three pathologists using 86 CRC HE slides. TSR and TILs were categorized using one and four different proposed cutoff systems, respectively, and agreement was assessed using the intraclass coefficient (ICC) and Cohen's kappa statistics. Pairwise evaluation of agreement was performed using the Fleiss kappa statistic and the concordance rate and it was visualized by Bland-Altman plots. To investigate the association between biomarkers and patient data, Pearson's correlation analysis was applied. RESULTS: For the evaluation of intratumoral stromal TILs, ICC of 0.505 (95% CI: 0.35-0.64) was obtained, kappa values were in the range of 0.21 to 0.38, and concordance rates in the range of 0.61 to 0.72. For the evaluation of tumor-front TILs, ICC was 0.52 (95% CI: 0.32-0.67), the overall kappa value ranged from 0.24 to 0.30, and the concordance rate ranged from 0.66 to 0.72. For estimating the TSR, the ICC was 0.48 (95% CI: 0.35-0.60), the kappa value was 0.49 and the concordance rate was 0.76. We observed a significant correlation between tumor grade and the median of TSR (0.29 (95% CI: 0.032-0.51), p-value = 0.03). CONCLUSIONS: The agreement between pathologists in estimating these markers corresponds to poor-to-moderate agreement; implementing immune scores in daily practice requires more concentration in inter-observer agreements.

Early prediction of medical students' performance in high-stakes examinations using machine learning approaches.

Introduction: Since the advent of medical education systems, managing high-stakes exams has been a top priority and challenge for all policymakers. However, considering machine learning (ML) techniques as a replacement for medical licensing examinations, particularly during crises such as the COVID-19 outbreak, could be an effective solution. This study uses ML models to develop a framework for predicting medical students' performance on high-stakes exams, such as the Comprehensive Medical Basic Sciences Examination (CMBSE). Material and methods: Prediction of students' status and score on high-stakes examinations faces several challenges, including an imbalanced number of failing and passing students, a large number of heterogeneous and complex features, and the need to identify at-risk and top-performing students. In this study, two major categories of ML approaches are compared: first, classic models (logistic regression (LR), support vector machine (SVM), and k-nearest neighbors (KNN)), and second, ensemble models (voting, bagging (BG), random forests (RF), adaptive boosting (ADA), extreme gradient boosting (XGB), and stacking). Results: To evaluate the models' discrimination ability, they are assessed using a real dataset containing information on medical students over a five-year period (n = 1005). The findings indicate that ensemble ML models demonstrate optimal performance in predicting CMBSE status (RF and stacking). Similarly, among the classic regressors, LR exhibited the highest root-mean-square deviation (RMSD) (0.134) and coefficient of determination (R2) (0.62), whereas the RF model had the highest RMSD (0.077) and R2 (0.80) overall. Furthermore, Anatomical Sciences, Biochemistry, Parasitology, and Entomology grade point average (GPA) and grades demonstrated the strongest positive correlation with the outcomes. Conclusion: Comparing classic and ensemble ML models revealed that ensemble models are superior to classic models. Therefore, the presented framework could be considered a suitable alternative for the CMBSE and other comparable medical licensing examinations.

Non-Alcoholic Fatty Liver Disease Diagnosis with Multi-Group Factors.

Although various clinical factors affect the diagnosis of Non-alcoholic Fatty Liver Disease (NAFLD), most studies only use single-source data such as images or laboratory data. Nevertheless, using different categories of features can help to get better results. Hence, one of the most important purposes of this paper is to employ a multi-group of effective factors such as velocimetry, psychological, demographic and anthropometric, and lab test data. Then, some Machine Learning (ML) methods are applied to classify the samples into two healthy and patient with NAFLD groups. The data used here belongs to the PERSIAN Organizational Cohort study at Mashhad University of Medical Sciences. To quantify the scalability of the models, different validity metrics are used. The obtained results illustrate that the proposed method can lead to an increase in the efficiency of the classifiers.