Decision trees: an overview and their use in medicine

Discovery of novel JAK1 inhibitors through combining machine learning, structure-based pharmacophore modeling and bio-evaluation

BACKGROUND

Janus kinase 1 (JAK1) plays a critical role in most cytokine-mediated inflammatory, autoimmune responses and various cancers via the JAK/STAT signaling pathway. Inhibition of JAK1 is therefore an attractive therapeutic strategy for several diseases. Recently, high-performance machine learning techniques have been increasingly applied in virtual screening to develop new kinase inhibitors. Our study aimed to develop a novel layered virtual screening method based on machine learning (ML) and pharmacophore models to identify the potential JAK1 inhibitors.

METHODS

Firstly, we constructed a high-quality dataset comprising 3834 JAK1 inhibitors and 12,230 decoys, followed by establishing a series of classification models based on a combination of three molecular descriptors and six ML algorithms. To further screen potential compounds, we constructed several pharmacophore models based on Hiphop and receptor-ligand algorithms. We then used molecular docking to filter the recognized compounds. Finally, the binding stability and enzyme inhibition activity of the identified compounds were assessed by molecular dynamics (MD) simulations and in vitro enzyme activity tests.

RESULTS

The best performance ML model DNN-ECFP4 and two pharmacophore models Hiphop3 and 6TPF 08 were utilized to screen the ZINC database. A total of 13 potentially active compounds were screened and the MD results demonstrated that all of the above molecules could bind with JAK1 stably in dynamic conditions. Among the shortlisted compounds, the four purchasable compounds demonstrated significant kinase inhibition activity, with Z-10 being the most active (IC50 = 194.9 nM).

CONCLUSION

The current study provides an efficient and accurate integrated model. The hit compounds were promising candidates for the further development of novel JAK1 inhibitors.

A Data-Driven Approach to Refine Predictions of Differentiated Thyroid Cancer Outcomes: A Prospective Multicenter Study

CONTEXT

The risk stratification of patients with differentiated thyroid cancer (DTC) is crucial in clinical decision making. The most widely accepted method to assess risk of recurrent/persistent disease is described in the 2015 American Thyroid Association (ATA) guidelines. However, recent research has focused on the inclusion of novel features or questioned the relevance of currently included features.

OBJECTIVE

To develop a comprehensive data-driven model to predict persistent/recurrent disease that can capture all available features and determine the weight of predictors.

METHODS

In a prospective cohort study, using the Italian Thyroid Cancer Observatory (ITCO) database (NCT04031339), we selected consecutive cases with DTC and at least early follow-up data (n = 4773; median follow-up 26 months; interquartile range, 12-46 months) at 40 Italian clinical centers. A decision tree was built to assign a risk index to each patient. The model allowed us to investigate the impact of different variables in risk prediction.

RESULTS

By ATA risk estimation, 2492 patients (52.2%) were classified as low, 1873 (39.2%) as intermediate, and 408 as high risk. The decision tree model outperformed the ATA risk stratification system: the sensitivity of high-risk classification for structural disease increased from 37% to 49%, and the negative predictive value for low-risk patients increased by 3%. Feature importance was estimated. Several variables not included in the ATA system significantly impacted the prediction of disease persistence/recurrence: age, body mass index, tumor size, sex, family history of thyroid cancer, surgical approach, presurgical cytology, and circumstances of the diagnosis.

CONCLUSION

Current risk stratification systems may be complemented by the inclusion of other variables in order to improve the prediction of treatment response. A complete dataset allows for more precise patient clustering.

Arterial stiffness and biological parameters: A decision tree machine learning application in hypertensive participants

Arterial stiffness, measured by arterial stiffness index (ASI), could be considered a main denominator in target organ damage among hypertensive subjects. Currently, no reported ASI normal references have been reported. The index of arterial stiffness is evaluated by calculation of a stiffness index. Predicted ASI can be estimated regardless to age, sex, mean blood pressure, and heart rate, to compose an individual stiffness index [(measured ASI-predicted ASI)/predicted ASI]. A stiffness index greater than zero defines arterial stiffness. Thus, the purpose of this study was 1) to determine determinants of stiffness index 2) to perform threshold values to discriminate stiffness index and then 3) to determine hierarchical associations of the determinants by performing a decision tree model among hypertensive participants without CV diseases. A study was conducted from 53,363 healthy participants in the UK Biobank survey to determine predicted ASI. Stiffness index was applied on 49,452 hypertensives without CV diseases to discriminate determinants of positive stiffness index (N = 22,453) from negative index (N = 26,999). The input variables for the models were clinical and biological parameters. The independent classifiers were ranked from the most sensitives: HDL cholesterol≤1.425 mmol/L, smoking pack years≥9.2pack-years, Phosphate≥1.172 mmol/L, to the most specifics: Cystatin c≤0.901 mg/L, Triglycerides≥1.487 mmol/L, Urate≥291.9 μmol/L, ALT≥22.13 U/L, AST≤32.5 U/L, Albumin≤45.92 g/L, Testosterone≥5.181 nmol/L. A decision tree model was performed to determine rules to highlight the different hierarchization and interactions between these classifiers with a higher performance than multiple logistic regression (p<0.001). The stiffness index could be an integrator of CV risk factors and participate in future CV risk management evaluations for preventive strategies. Decision trees can provide accurate and useful classification for clinicians.

Rapid protocols to support COVID-19 clinical diagnosis based on hematological parameters

Purpose

In December 2019, the Covid-19 pandemic began in the world. To reduce mortality, in addiction to mass vaccination, it is necessary to massify and accelerate clinical diagnosis, as well as creating new ways of monitoring patients that can help in the construction of specific treatments for the disease.

Objective

In this work, we propose rapid protocols for clinical diagnosis of COVID-19 through the automatic analysis of hematological parameters using evolutionary computing and machine learning. These hematological parameters are obtained from blood tests common in clinical practice.

Method

We investigated the best classifier architectures. Then, we applied the particle swarm optimization algorithm (PSO) to select the most relevant attributes: serum glucose, troponin, partial thromboplastin time, ferritin, D-dimer, lactic dehydrogenase, and indirect bilirubin. Then, we assessed again the best classifier architectures, but now using the reduced set of features. Finally, we used decision trees to build four rapid protocols for Covid-19 clinical diagnosis by assessing the impact of each selected feature. The proposed system was used to support clinical diagnosis and assessment of disease severity in patients admitted to intensive and semi-intensive care units as a case study in the city of Paudalho, Brazil.

Results

We developed a web system for Covid-19 diagnosis support. Using a 100-tree random forest, we obtained results for accuracy, sensitivity, and specificity superior to 99%. After feature selection, results were similar. The four empirical clinical protocols returned accuracies, sensitivities and specificities superior to 98%.

Conclusion

By using a reduced set of hematological parameters common in clinical practice, it was possible to achieve results of accuracy, sensitivity, and specificity comparable to those obtained with RT-PCR. It was also possible to automatically generate clinical decision protocols, allowing relatively accurate clinical diagnosis even without the aid of the web decision support system.

Machine learning-based predictive model for prevention of metabolic syndrome

Metabolic syndrome (MetS) is a chronic disease caused by obesity, high blood pressure, high blood sugar, and dyslipidemia and may lead to cardiovascular disease or type 2 diabetes. Therefore, the detection and prevention of MetS at an early stage are imperative. Individuals can detect MetS early and manage it effectively if they can easily monitor their health status in their daily lives. In this study, a predictive model for MetS was developed utilizing solely noninvasive information, thereby facilitating its practical application in real-world scenarios. The model's construction deliberately excluded three features requiring blood testing, specifically those for triglycerides, blood sugar, and HDL cholesterol. We used a large-scale Korean health examination dataset (n = 70, 370; the prevalence of MetS = 13.6%) to develop the predictive model. To obtain informative features, we developed three novel synthetic features from four basic information: waist circumference, systolic and diastolic blood pressure, and gender. We tested several classification algorithms and confirmed that the decision tree model is the most appropriate for the practical prediction of MetS. The proposed model achieved good performance, with an AUC of 0.889, a recall of 0.855, and a specificity of 0.773. It uses only four base features, which results in simplicity and easy interpretability of the model. In addition, we performed calibrations on the prediction probability and calibrated the model. Therefore, the proposed model can provide MetS diagnosis and risk prediction results. We also proposed a MetS risk map such that individuals could easily determine whether they had metabolic syndrome.

Using a decision tree algorithm to distinguish between repeated supra-therapeutic and acute acetaminophen exposures

BACKGROUND

This study aimed to compare clinical and laboratory characteristics of supra-therapeutic (RSTI) and acute acetaminophen exposures using a predictive decision tree (DT) algorithm.

METHODS

We conducted a retrospective cohort study using the National Poison Data System (NPDS). All patients with RSTI acetaminophen exposure (n = 4,522) between January 2012 and December 2017 were included. Additionally, 4,522 randomly selected acute acetaminophen ingestion cases were included. After that, the DT machine learning algorithm was applied to differentiate acute acetaminophen exposure from supratherapeutic exposures.

RESULTS

The DT model had accuracy, precision, recall, and F1-scores of 0.75, respectively. Age was the most relevant variable in predicting the type of acetaminophen exposure, whether RSTI or acute. Serum aminotransferase concentrations, abdominal pain, drowsiness/lethargy, and nausea/vomiting were the other most important factors distinguishing between RST and acute acetaminophen exposure.

CONCLUSION

DT models can potentially aid in distinguishing between acute and RSTI of acetaminophen. Further validation is needed to assess the clinical utility of this model.

Decision Tree Analyses for Prediction of QoL over a One-Year Period in Breast Cancer Patients: An Added Value of Patient-Reported Outcomes

Despite the current shift in medicine towards patient-centered care, clinicians rarely utilize patient-reported outcomes (PROs) in everyday practice. We examined the predictors of quality- of-life (QoL) trajectories in breast cancer (BC) patients during the first year after primary treatment. A total of 185 BC patients referred for postoperative radiotherapy (RT) filled in the EORTC QLQ-C30 Questionnaire assessing global QoL, functioning and cancer-related symptoms before starting RT; directly after RT; and 3, 6 and 12 months after RT. We used decision tree analyses to examine which baseline factors best allowed for predicting the one-year trajectory of the global QoL after BC treatment. We tested two models: 'basic', including medical and sociodemographic characteristics, and 'enriched', additionally including PROs. We recognized three distinct trajectories of global QoL: 'high', 'U-shape' and 'low'. Of the two compared models, the 'enriched' model allowed for a more accurate prediction of a given QoL trajectory, with all indicators of model validation being better. In this model, baseline global QoL and functioning measures were the key discriminators of QoL trajectory. Taking PROs into account increases the accuracy of the prediction model. Collecting this information in the clinical interview is recommended, especially for patients with lower QoL.

Quantitative Electroencephalography Analysis for Improved Assessment of Consciousness Levels in Deep Coma Patients Using a Proposed Stimulus Stage

"Coma" is defined as an inability to obey commands, to speak, or to open the eyes. So, a coma is a state of unarousable unconsciousness. In a clinical setting, the ability to respond to a command is often used to infer consciousness. Evaluation of the patient's level of consciousness (LeOC) is important for neurological evaluation. The Glasgow Coma Scale (GCS) is the most widely used and popular scoring system for neurological evaluation and is used to assess a patient's level of consciousness. The aim of this study is the evaluation of GCSs with an objective approach based on numerical results. So, EEG signals were recorded from 39 patients in a coma state with a new procedure proposed by us in a deep coma state (GCS: between 3 and 8). The EEG signals were divided into four sub-bands as alpha, beta, delta, and theta, and their power spectral density was calculated. As a result of power spectral analysis, 10 different features were extracted from EEG signals in the time and frequency domains. The features were statistically analyzed to differentiate the different LeOC and to relate with the GCS. Additionally, some machine learning algorithms have been used to measure the performance of the features for distinguishing patients with different GCSs in a deep coma. This study demonstrated that GCS 3 and GCS 8 patients were classified from other levels of consciousness in terms of decreased theta activity. To the best of our knowledge, this is the first study to classify patients in a deep coma (GCS between 3 and 8) with 96.44% classification performance.

Comparison of decision tree with common machine learning models for prediction of biguanide and sulfonylurea poisoning in the United States: an analysis of the National Poison Data System

BACKGROUND

Biguanides and sulfonylurea are two classes of anti-diabetic medications that have commonly been prescribed all around the world. Diagnosis of biguanide and sulfonylurea exposures is based on history taking and physical examination; thus, physicians might misdiagnose these two different clinical settings. We aimed to conduct a study to develop a model based on decision tree analysis to help physicians better diagnose these poisoning cases.

METHODS

The National Poison Data System was used for this six-year retrospective cohort study.The decision tree model, common machine learning models multi layers perceptron, stochastic gradient descent (SGD), Adaboosting classiefier, linear support vector machine and ensembling methods including bagging, voting and stacking methods were used. The confusion matrix, precision, recall, specificity, f1-score, and accuracy were reported to evaluate the model's performance.

RESULTS

Of 6183 participants, 3336 patients (54.0%) were identified as biguanides exposures, and the remaining were those with sulfonylureas exposures. The decision tree model showed that the most important clinical findings defining biguanide and sulfonylurea exposures were hypoglycemia, abdominal pain, acidosis, diaphoresis, tremor, vomiting, diarrhea, age, and reasons for exposure. The specificity, precision, recall, f1-score, and accuracy of all models were greater than 86%, 89%, 88%, and 88%, respectively. The lowest values belong to SGD model. The decision tree model has a sensitivity (recall) of 93.3%, specificity of 92.8%, precision of 93.4%, f1_score of 93.3%, and accuracy of 93.3%.

CONCLUSION

Our results indicated that machine learning methods including decision tree and ensembling methods provide a precise prediction model to diagnose biguanides and sulfonylureas exposure.

Predicting heart failure onset in the general population using a novel data-mining artificial intelligence method

We aimed to identify combinations of clinical factors that predict heart failure (HF) onset using a novel limitless-arity multiple-testing procedure (LAMP). We also determined if increases in numbers of predictive combinations of factors increases the probability of developing HF. We recruited people without HF who received health check-ups in 2010, who were followed annually for 4 years. Using 32,547 people, LAMP was performed to identify combinations of factors of fewer than four factors that could predict the onset of HF. The ability of the method to predict the probability of HF onset based on the number of matching predictive combinations of factors was determined in 275,658 people. We identified 549 combinations of factors for the onset of HF. Then we classified 275,658 people into six groups who had 0, 1-50, 51-100, 101-150, 151-200 or 201-250 predictive combinations of factors for the onset of HF. We found that the probability of HF progressively increased as the number of predictive combinations of factors increased. We identified combinations of variables that predict HF onset. An increased number of matching predictive combinations for the onset of HF increased the probability of HF onset.

Fusion of pattern-based and statistical features for Schizophrenia detection from EEG signals

Schizophrenia (SZ) is a chronic disorder affecting the functioning of the brain. It can lead to irrational behaviour amongst the patients suffering from this disease. A low-cost diagnostic needs to be developed for SZ so that timely treatment can be provided to the patients. In this work, we propose an accurate and easy-to-implement system to detect SZ using electroencephalogram (EEG) signals. The signal is divided into sub-band components by a Fourier-based technique that can be implemented in real-time using fast Fourier transform. Thereafter, statistical features are computed from these components. Further, look ahead pattern (LAP) is developed as a feature to capture local variations in the EEG signal. The fusion of these two distinct schemes enables a thorough examination of EEG signals. Kruskal-Wallis test is utilized for the selection of significant features. Various machine learning classifiers are employed and the proposed framework achieves 98.62% and 99.24% accuracy in identifying SZ cases, considering two distinct datasets, using boosted trees classifier. This method provides a promising candidate for widespread deployment in efficient real-time systems for SZ detection.

Proteochemometrics modeling for prediction of the interactions between caspase isoforms and their inhibitors

Caspases (cysteine-aspartic proteases) play critical roles in inflammation and the programming of cell death in the form of necroptosis, apoptosis, and pyroptosis. The name of these enzymes has been chosen in accordance with their cysteine protease activity. They act as cysteines in nucleophilically active sites to attack and cleave target proteins in the aspartic acid and amino acid C-terminal. Based on the substrate's structure and the specificity, the physiological activity of caspases is divided. However, in apoptosis, the division of caspases into initiating caspases (caspase 2, 8, 9, and 10) and executive caspases (caspase 3, 6, and 7) is essential. The present study aimed to perform Proteochemometrics Modeling to generalize the data on caspases, which could predict ligand and protein interactions. In this study, we employed protein and ligand descriptors. Moreover, protein descriptors were computed using the Protr R package, while PADEL-Descriptor was employed for the computation of ligand descriptors. In addition, NCA (Neighborhood Component Analyses) was used for descriptor selection, and SVR, decision tree, and ensemble methods were utilized for the proteochemometrics modeling. This study shows that the ensemble model demonstrates superior performance compared with other models in terms of R², Q², and RMSE criteria.

Predictors of spinal trauma care and outcomes in a resource-constrained environment: a decision tree analysis of spinal trauma surgery and outcomes in Tanzania

OBJECTIVE

The burden of spinal trauma in low- and middle-income countries (LMICs) is immense, and its management is made complex in such resource-restricted settings. Algorithmic evidence-based management is cost-prohibitive, especially with respect to spinal implants, while perioperative care is work-intensive, making overall care dependent on multiple constraints. The objective of this study was to identify determinants of decision-making for surgical intervention, improvement in function, and in-hospital mortality among patients experiencing acute spinal trauma in resource-constrained settings.

METHODS

This study was a retrospective analysis of prospectively collected data in a cohort of patients with spinal trauma admitted to a tertiary referral hospital center in Dar es Salam, Tanzania. Data on demographic, clinical, and treatment characteristics were collected as part of a quality improvement neurotrauma registry. Outcome measures were surgical intervention, American Spinal Injury Association (ASIA) Impairment Scale (AIS) grade improvement, and in-hospital mortality, based on existing treatment protocols. Univariate analyses of demographic and clinical characteristics were performed for each outcome of interest. Using the variables associated with each outcome, a machine learning algorithm-based regression nonparametric decision tree model utilizing a bootstrapping method was created and the accuracy of the three models was estimated.

RESULTS

Two hundred eighty-four consecutively admitted patients with acute spinal trauma were included over a period of 33 months. The median age was 34 (IQR 26-43) years, 83.8% were male, and 50.7% had experienced injury in a motor vehicle accident. The median time to hospital admission after injury was 2 (IQR 1-6) days; surgery was performed after a further median delay of 22 (IQR 13-39) days. Cervical spine injury comprised 38.4% of the injuries. Admission AIS grades were A in 48.9%, B in 16.2%, C in 8.5%, D in 9.5%, and E in 16.6%. Nearly half (45.1%) of the patients underwent surgery, 12% had at least one functional improvement in AIS grade, and 11.6% died in the hospital. Determinants of surgical intervention were age ≤ 30 years, spinal injury level, admission AIS grade, delay in arrival to the referral hospital, undergoing MRI, and type of insurance; admission AIS grade, delay to arrival to the hospital, and injury level for functional improvement; and delay to arrival, injury level, delay to surgery, and admission AIS grade for in-hospital mortality. The best accuracies for the decision tree models were 0.62, 0.34, and 0.93 for surgery, AIS grade improvement, and in-hospital mortality, respectively.

CONCLUSIONS

Operative intervention and functional improvement after acute spinal trauma in this tertiary referral hospital in an LMIC environment were low and inconsistent, which suggests that nonclinical factors exist within complex resource-driven decision-making frameworks. These nonclinical factors are highlighted by the authors' results showing clinical outcomes and in-hospital mortality were determined by natural history, as evidenced by the highest accuracy of the model predicting in-hospital mortality.

Prediction of left ventricular ejection fraction changes in heart failure patients using machine learning and electronic health records: a multi-site study

Left ventricular ejection fraction (EF) is a key measure in the diagnosis and treatment of heart failure (HF) and many patients experience changes in EF overtime. Large-scale analysis of longitudinal changes in EF using electronic health records (EHRs) is limited. In a multi-site retrospective study using EHR data from three academic medical centers, we investigated longitudinal changes in EF measurements in patients diagnosed with HF. We observed significant variations in baseline characteristics and longitudinal EF change behavior of the HF cohorts from a previous study that is based on HF registry data. Data gathered from this longitudinal study were used to develop multiple machine learning models to predict changes in ejection fraction measurements in HF patients. Across all three sites, we observed higher performance in predicting EF increase over a 1-year duration, with similarly higher performance predicting an EF increase of 30% from baseline compared to lower percentage increases. In predicting EF decrease we found moderate to high performance with low confidence for various models. Among various machine learning models, XGBoost was the best performing model for predicting EF changes. Across the three sites, the XGBoost model had an F1-score of 87.2, 89.9, and 88.6 and AUC of 0.83, 0.87, and 0.90 in predicting a 30% increase in EF, and had an F1-score of 95.0, 90.6, 90.1 and AUC of 0.54, 0.56, 0.68 in predicting a 30% decrease in EF. Among features that contribute to predicting EF changes, baseline ejection fraction measurement, age, gender, and heart diseases were found to be statistically significant.

AI-Powered Blockchain Technology for Public Health: A Contemporary Review, Open Challenges, and Future Research Directions

Blockchain technology has been growing at a substantial growth rate over the last decade. Introduced as the backbone of cryptocurrencies such as Bitcoin, it soon found its application in other fields because of its security and privacy features. Blockchain has been used in the healthcare industry for several purposes including secure data logging, transactions, and maintenance using smart contracts. Great work has been carried out to make blockchain smart, with the integration of Artificial Intelligence (AI) to combine the best features of the two technologies. This review incorporates the conceptual and functional aspects of the individual technologies and innovations in the domains of blockchain and artificial intelligence and lays down a strong foundational understanding of the domains individually and also rigorously discusses the various ways AI has been used along with blockchain to power the healthcare industry including areas of great importance such as electronic health record (EHR) management, distant-patient monitoring and telemedicine, genomics, drug research, and testing, specialized imaging and outbreak prediction. It compiles various algorithms from supervised and unsupervised machine learning problems along with deep learning algorithms such as convolutional/recurrent neural networks and numerous platforms currently being used in AI-powered blockchain systems and discusses their applications. The review also presents the challenges still faced by these systems which they inherit from the AI and blockchain algorithms used at the core of them and the scope of future work.

Covid-19 rapid test by combining a Random Forest-based web system and blood tests

The disease caused by the new type of coronavirus, Covid-19, has posed major public health challenges for many countries. With its rapid spread, since the beginning of the outbreak in December 2019, the disease transmitted by SARS-CoV-2 has already caused over 2 million deaths to date. In this work, we propose a web solution, called Heg.IA, to optimize the diagnosis of Covid-19 through the use of artificial intelligence. Our system aims to support decision-making regarding to diagnosis of Covid-19 and to the indication of hospitalization on regular ward, semi-ICU or ICU based on decision a Random Forest architecture with 90 trees. The main idea is that healthcare professionals can insert 41 hematological parameters from common blood tests and arterial gasometry into the system. Then, Heg.IA will provide a diagnostic report. The system reached good results for both Covid-19 diagnosis and to recommend hospitalization. For the first scenario we found average results of accuracy of 92.891%±0.851, kappa index of 0.858 ± 0.017, sensitivity of 0.936 ± 0.011, precision of 0.923 ± 0.011, specificity of 0.921 ± 0.012 and area under ROC of 0.984 ± 0.003. As for the indication of hospitalization, we achieved excellent performance of accuracies above 99% and more than 0.99 for the other metrics in all situations. By using a computationally simple method, based on the classical decision trees, we were able to achieve high diagnosis performance. Heg.IA system may be a way to overcome the testing unavailability in the context of Covid-19.Communicated by Ramaswamy H. Sarma.

Identifying hepatocellular carcinoma patients with survival benefits from surgery combined with chemotherapy: based on machine learning model

BACKGROUND

Hepatocellular carcinoma (HCC) is still fatal even after surgical resection. The purpose of this study was to analyze the prognostic factors of 5-year survival rate and to establish a model to identify HCC patients with gain of surgery combined with chemotherapy.

METHODS

All patients with HCC after surgery from January 2010 to December 2015 were selected from the Surveillance, Epidemiology, and End Results (SEER) database. Univariate and multivariate logistic analysis were used to analyze the prognostic factors of patients, and the risk prediction model of 5-year survival rate of HCC patients was established by classical decision tree method. Propensity score matching was used to eliminate the confounding factors of whether to receive chemotherapy in high-risk group or low-risk group.

RESULTS

One-thousand six-hundred twenty-five eligible HCC patients were included in the study. Marital status, α-fetoprotein (AFP), vascular infiltration, tumor size, number of lesions, and grade were independent prognostic factors affecting the 5-year survival rate of HCC patients. The area under the curve of the 5-year survival risk prediction model constructed from the above variables was 0.76, and the classification accuracy, precision, recall, and F1 scores were 0.752, 0.83, 0.842, and 0.836, respectively. High-risk patients classified according to the prediction model had better 5-year survival rate after chemotherapy, while there was no difference in 5-year survival rate between patients receiving chemotherapy and patients not receiving chemotherapy in the low-risk group.

CONCLUSIONS

The 5-year survival risk prediction model constructed in this study provides accurate survival prediction information. The high-risk patients determined according to the prediction model may benefit from the 5-year survival rate after combined chemotherapy.

Machine learning models predicting undertriage in telephone triage

BACKGROUND

Undertriaged patients have worse outcomes than appropriately triaged patients. Machine learning provides better triage prediction than conventional triage in emergency departments, but no machine learning-based undertriage prediction models have yet been developed for prehospital telephone triage. We developed and validated machine learning models for telephone triage.

MATERIALS AND METHODS

We conducted a retrospective cohort study with the largest after-hour house-call (AHHC) service dataset in Japan. Participants were ≥16 years and used the AHHC service between 1 November 2018 and 31 January 2021. We developed five prediction models based on support vector machine (SVM), lasso regression (LR), random forest (RF), gradient-boosted decision tree (XGB), and deep neural network (DNN). The primary outcome was undertriage, and predictors were telephone triage level and routinely available telephone-based data, including age, sex, 80 chief complaint categories and 10 comorbidities. We measured the area under the receiver operating characteristic curve (AUROC) for all the models.

RESULTS

We identified 15,442 eligible patients (age: 38.4 ± 16.6, male: 57.2%), including 298 (1.9%; age: 58.2 ± 23.9, male: 55.0%) undertriaged patients. RF and XGB outperformed the other models, with the AUROC values (95% confidence interval; 95% CI) of the SVM, LR, RF, XGB and DNN for undertriage being 0.62 (0.55-0.69), 0.79 (0.74-0.83), 0.81 (0.76-0.86), 0.80 (0.75-0.84) and 0.77 (0.73-0.82), respectively.

CONCLUSIONS

We found that RF and XGB outperformed other models. Our findings suggest that machine learning models can facilitate the early detection of undertriage and early intervention to yield substantially improved patient outcomes.KEY MESSAGESUndertriaged patients experience worse outcomes than appropriately triaged patients; thus, we developed machine learning models for predicting undertriage in the prehospital setting. In addition, we identified the predictors of risk factors associated with undertriage.Random forest and gradient-boosted decision tree models demonstrated better prediction performance, and the models identified the risk factors associated with undertriage.Machine learning models aid in the early detection of undertriage, leading to significantly improved patient outcomes and identifying undertriage-associated risk factors, including chief complaint categories, could help prioritize conventional telephone triage protocol revision.

Forecasting determinants of recurrence in lung cancer patients exploiting various machine learning models

Lung cancer recurrence seems to be the most leading cause of death as well as deterioration of lifespan. Proper assessment of the probability of recurrence in early-stage lung cancer is necessary to push up the treatment progress. We therefore employed machine-learning technologies to forecast post-operative recurrence risks using 174 lung cancer patient records. Six classification algorithms logistic regression, SVM, decision tree classification, random forest classification, XGBoost and lightGBM were used to predict the cancer recurrence. The patient samples were divided into training and test group with the split ratio of 3:1 for model generation and the accuracy were validated using k-fold cross-validation method. It is worth noting that the logistic regression model outperformed all the models in both training (Accuracy = 0.82) and test set (Accuracy = 0.79) on k-fold validation. Further, the optimal features (n = 7) identified using the RFE method is certainly helpful to improve the model in a high precision. The imperative risk factors associated with recurrence were identified using three feature selection methods. Importantly, our research showed that age is an important prognostic factor to be considered during the recurrence prediction. Indeed, severe concern on the identified risk factors combined with predictive models assists the physician to reduce the cancer recurrence rate in patients with lung cancer.

Prediction of radiographic progression pattern in patients with ankylosing spondylitis using group-based trajectory modeling and decision trees

Objective

This study aimed to identify trajectories of radiographic progression of the spine over time and use them, along with associated clinical factors, to develop a prediction model for patients with ankylosing spondylitis (AS).

Methods

Data from the medical records of patients diagnosed with AS in a single center were extracted between 2001 and 2018. Modified Stoke Ankylosing Spondylitis Spinal Scores (mSASSS) were estimated from cervical and lumbar radiographs. Group-based trajectory modeling classified patients into trajectory subgroups using longitudinal mSASSS data. In multivariate analysis, significant clinical factors associated with trajectories were selected and used to develop a decision tree for prediction of radiographic progression. The most appropriate group for each patient was then predicted using decision tree analysis.

Results

We identified three trajectory classes: class 1 had a uniformly increasing slope of mSASSS, class 2 showed sustained low mSASSS, and class 3 showed little change in the slope of mSASSS but highest mSASSS from time of diagnosis to after progression. In multivariate analysis for predictive factors, female sex, younger age at diagnosis, lack of eye involvement, presence of peripheral joint involvement, and low baseline erythrocyte sedimentation rate (log) were significantly associated with class 2. Class 3 was significantly associated with male sex, older age at diagnosis, presence of ocular involvement, and lack of peripheral joint involvement when compared with class 1. Six clinical factors from multivariate analysis were used for the decision tree for classifying patients into three trajectories of radiographic progression.

Conclusion

We identified three patterns of radiographic progression over time and developed a decision tree based on clinical factors to classify patients according to their trajectories of radiographic progression. Clinically, this model holds promise for predicting prognosis in patients with AS.

Task-Based Visual Interactive Modeling: Decision Trees and Rule-Based Classifiers

Visual analytics enables the coupling of machine learning models and humans in a tightly integrated workflow, addressing various analysis tasks. Each task poses distinct demands to analysts and decision-makers. In this survey, we focus on one canonical technique for rule-based classification, namely decision tree classifiers. We provide an overview of available visualizations for decision trees with a focus on how visualizations differ with respect to 16 tasks. Further, we investigate the types of visual designs employed, and the quality measures presented. We find that (i) interactive visual analytics systems for classifier development offer a variety of visual designs, (ii) utilization tasks are sparsely covered, (iii) beyond classifier development, node-link diagrams are omnipresent, (iv) even systems designed for machine learning experts rarely feature visual representations of quality measures other than accuracy. In conclusion, we see a potential for integrating algorithmic techniques, mathematical quality measures, and tailored interactive visualizations to enable human experts to utilize their knowledge more effectively.

Qualitative Evaluation of an Artificial Intelligence–Based Clinical Decision Support System to Guide Rhythm Management of Atrial Fibrillation: Survey Study

Background

Despite the numerous studies evaluating various rhythm control strategies for atrial fibrillation (AF), determination of the optimal strategy in a single patient is often based on trial and error, with no one-size-fits-all approach based on international guidelines/recommendations. The decision, therefore, remains personal and lends itself well to help from a clinical decision support system, specifically one guided by artificial intelligence (AI). QRhythm utilizes a 2-stage machine learning (ML) model to identify the optimal rhythm management strategy in a given patient based on a set of clinical factors, in which the model first uses supervised learning to predict the actions of an expert clinician and identifies the best strategy through reinforcement learning to obtain the best clinical outcome—a composite of symptomatic recurrence, hospitalization, and stroke.

Objective

We qualitatively evaluated a novel, AI-based, clinical decision support system (CDSS) for AF rhythm management, called QRhythm, which uses both supervised and reinforcement learning to recommend either a rate control or one of 3 types of rhythm control strategies—external cardioversion, antiarrhythmic medication, or ablation—based on individual patient characteristics.

Methods

Thirty-three clinicians, including cardiology attendings and fellows and internal medicine attendings and residents, performed an assessment of QRhythm, followed by a survey to assess relative comfort with automated CDSS in rhythm management and to examine areas for future development.

Results

The 33 providers were surveyed with training levels ranging from resident to fellow to attending. Of the characteristics of the app surveyed, safety was most important to providers, with an average importance rating of 4.7 out of 5 (SD 0.72). This priority was followed by clinical integrity (a desire for the advice provided to make clinical sense; importance rating 4.5, SD 0.9), backward interpretability (transparency in the population used to create the algorithm; importance rating 4.3, SD 0.65), transparency of the algorithm (reasoning underlying the decisions made; importance rating 4.3, SD 0.88), and provider autonomy (the ability to challenge the decisions made by the model; importance rating 3.85, SD 0.83). Providers who used the app ranked the integrity of recommendations as their highest concern with ongoing clinical use of the model, followed by efficacy of the application and patient data security. Trust in the app varied; 1 (17%) provider responded that they somewhat disagreed with the statement, “I trust the recommendations provided by the QRhythm app,” 2 (33%) providers responded with neutrality to the statement, and 3 (50%) somewhat agreed with the statement.

Conclusions

Safety of ML applications was the highest priority of the providers surveyed, and trust of such models remains varied. Widespread clinical acceptance of ML in health care is dependent on how much providers trust the algorithms. Building this trust involves ensuring transparency and interpretability of the model.

Care-seeking delay of imported malaria to China: implications for improving post-travel healthcare for migrant workers

BACKGROUND

Imported malaria cases continue to pose major challenges in China as well as in other countries having achieved elimination. Our study aims to identify the factors influencing the timing of care-seeking after symptom onset among migrant workers with imported malaria, in order to develop innovative interventions to improve access and provision of post-travel healthcare for returning migrants.

METHODS

We analysed the timing and types of healthcare service utilization after symptom onset among patients with imported malaria between 2012 and 2019 in Jiangsu Province, China. Moreover, decision tree models were used to explore the factors influencing the care-seeking timing after symptom onset among patients with imported malaria.

RESULTS

A total of 2255 cases of imported malaria were identified from 1 June 2012 through 31 December 2019. Patients with malaria imported into China were mainly male migrant labourers returning from sub-Saharan Africa (96.8%). A substantial number of patients with imported malaria sought healthcare >3 days after symptom onset, which clearly represented delayed healthcare-seeking behaviour. According to the decision tree analysis, initial healthcare seeking from healthcare facilities at higher administrative levels, infection with Plasmodium vivax and absence of malaria infection history were significantly associated with delayed healthcare seeking in patients with imported malaria.

CONCLUSION

The delay in seeking of medical care among migrant workers with imported malaria should be considered and addressed by specific interventions. In addition to increasing awareness about these issues among health care professionals, improved access to healthcare facilities at higher administrative levels as well as improved diagnostic capacity of healthcare facilities at lower administrative levels should be developed. Moreover, education programs targeting populations at risk of malaria importation and delayed healthcare seeking should be improved to facilitate early healthcare seeking and service use.

Utility of support vector machine and decision tree to identify the prognosis of metformin poisoning in the United States: analysis of National Poisoning Data System

BACKGROUND

With diabetes incidence growing globally and metformin still being the first-line for its treatment, metformin's toxicity and overdose have been increasing. Hence, its mortality rate is increasing. For the first time, we aimed to study the efficacy of machine learning algorithms in predicting the outcome of metformin poisoning using two well-known classification methods, including support vector machine (SVM) and decision tree (DT).

METHODS

This study is a retrospective cohort study of National Poison Data System (NPDS) data, the largest data repository of poisoning cases in the United States. The SVM and DT algorithms were developed using training and test datasets. We also used precision-recall and ROC curves and Area Under the Curve value (AUC) for model evaluation.

RESULTS

Our model showed that acidosis, hypoglycemia, electrolyte abnormality, hypotension, elevated anion gap, elevated creatinine, tachycardia, and renal failure are the most important determinants in terms of outcome prediction of metformin poisoning. The average negative predictive value for the decision tree and SVM models was 92.30 and 93.30. The AUC of the ROC curve of the decision tree for major, minor, and moderate outcomes was 0.92, 0.92, and 0.89, respectively. While this figure of SVM model for major, minor, and moderate outcomes was 0.98, 0.90, and 0.82, respectively.

CONCLUSIONS

In order to predict the prognosis of metformin poisoning, machine learning algorithms might help clinicians in the management and follow-up of metformin poisoning cases.

Assessment of the Perception of Sustainability for Occupants of Residential Buildings: A Case Study in the UAE

The residential sector is multi-faceted by nature. Although evidence shows that the UAE is among the countries in the world that take sustainability seriously, there is a lack of information about the perception of sustainability by occupants in the residential sector in the UAE. The aim of this paper is to assess the perception of sustainability of the residential sector in the UAE, which is achieved by following a methodological framework using the relevant literature review and experts’ knowledge. An online survey was distributed to the targeted population, followed by a statistical analysis to fulfill the aim of the paper. Results confirm the correlation between social, economic, and environmental aspects of sustainability. Additionally, structural equation modeling reveals that the perception of sustainability is significantly influenced by economic and environmental aspects in the residential sector in the UAE. Comparative analysis shows a statistical difference in the perception of sustainability among gender, educational level, employment status, and monthly income. Finally, a predictive classification model is built to classify the perception of occupants based on their attributes using decision tree algorithms. The outcomes of this study would be beneficial to policy and decision makers, developers, contractors, designers, and facility management entities to enhance overall sustainability in the residential sector.

Predicting the 2-Year Risk of Progression from Prediabetes to Diabetes Using Machine Learning among Chinese Elderly Adults

Identifying people with a high risk of developing diabetes among those with prediabetes may facilitate the implementation of a targeted lifestyle and pharmacological interventions. We aimed to establish machine learning models based on demographic and clinical characteristics to predict the risk of incident diabetes. We used data from the free medical examination service project for elderly people who were 65 years or older to develop logistic regression (LR), decision tree (DT), random forest (RF), and extreme gradient boosting (XGBoost) machine learning models for the follow-up results of 2019 and 2020 and performed internal validation. The receiver operating characteristic (ROC), sensitivity, specificity, accuracy, and F1 score were used to select the model with better performance. The average annual progression rate to diabetes in prediabetic elderly people was 14.21%. Each model was trained using eight features and one outcome variable from 9607 prediabetic individuals, and the performance of the models was assessed in 2402 prediabetes patients. The predictive ability of four models in the first year was better than in the second year. The XGBoost model performed relatively efficiently (ROC: 0.6742 for 2019 and 0.6707 for 2020). We established and compared four machine learning models to predict the risk of progression from prediabetes to diabetes. Although there was little difference in the performance of the four models, the XGBoost model had a relatively good ROC value, which might perform well in future exploration in this field.

Diagnosis and Prognosis of COVID-19 Disease Using Routine Blood Values and LogNNet Neural Network

Since February 2020, the world has been engaged in an intense struggle with the COVID-19 disease, and health systems have come under tragic pressure as the disease turned into a pandemic. The aim of this study is to obtain the most effective routine blood values (RBV) in the diagnosis and prognosis of COVID-19 using a backward feature elimination algorithm for the LogNNet reservoir neural network. The first dataset in the study consists of a total of 5296 patients with the same number of negative and positive COVID-19 tests. The LogNNet-model achieved the accuracy rate of 99.5% in the diagnosis of the disease with 46 features and the accuracy of 99.17% with only mean corpuscular hemoglobin concentration, mean corpuscular hemoglobin, and activated partial prothrombin time. The second dataset consists of a total of 3899 patients with a diagnosis of COVID-19 who were treated in hospital, of which 203 were severe patients and 3696 were mild patients. The model reached the accuracy rate of 94.4% in determining the prognosis of the disease with 48 features and the accuracy of 82.7% with only erythrocyte sedimentation rate, neutrophil count, and C reactive protein features. Our method will reduce the negative pressures on the health sector and help doctors to understand the pathogenesis of COVID-19 using the key features. The method is promising to create mobile health monitoring systems in the Internet of Things.

A machine-learning-based prediction method for easy COPD classification based on pulse oximetry clinical use

Chronic Obstructive Pulmonary Disease (COPD) is a progressive, obstructive lung disease that restricts airflow from the lungs. COPD patients are at risk of sudden and acute worsening of symptoms called exacerbations. Early identification and classification of COPD exacerbation can reduce COPD risks and improve patient’s healthcare and management. Pulse oximetry is a non-invasive technique used to assess patients with acutely worsening symptoms. As part of manual diagnosis based on pulse oximetry, clinicians examine three warning signs to classify COPD patients. This may lack high sensitivity and specificity which requires a blood test. However, laboratory tests require time, further delayed treatment and additional costs. This research proposes a prediction method for COPD patients’ classification based on pulse oximetry three manual warning signs and the resulting derived few key features that can be obtained in a short time. The model was developed on a robust physician labeled dataset with clinically diverse patient cases. Five classification algorithms were applied on the mentioned dataset and the results showed that the best algorithm is XGBoost with the accuracy of 91.04%, precision of 99.86%, recall of 82.19%, F1 measure value of 90.05% with an AUC value of 95.8%. Age, current and baseline heart rate, current and baseline pulse ox. (SPO2) were found the top most important predictors. These findings suggest the strength of XGBoost model together with the availability and the simplicity of input variables in classifying COPD daily living using a (wearable) pulse oximeter.

Diabetes mellitus prediction and diagnosis from a data preprocessing and machine learning perspective

BACKGROUND AND OBJECTIVE

Diabetes mellitus is a metabolic disorder characterized by hyperglycemia, which results from the inadequacy of the body to secrete and respond to insulin. If not properly managed or diagnosed on time, diabetes can pose a risk to vital body organs such as the eyes, kidneys, nerves, heart, and blood vessels and so can be life-threatening. The many years of research in computational diagnosis of diabetes have pointed to machine learning to as a viable solution for the prediction of diabetes. However, the accuracy rate to date suggests that there is still much room for improvement. In this paper, we are proposing a machine learning framework for diabetes prediction and diagnosis using the PIMA Indian dataset and the laboratory of the Medical City Hospital (LMCH) diabetes dataset. We hypothesize that adopting feature selection and missing value imputation methods can scale up the performance of classification models in diabetes prediction and diagnosis.

METHODS

In this paper, a robust framework for building a diabetes prediction model to aid in the clinical diagnosis of diabetes is proposed. The framework includes the adoption of Spearman correlation and polynomial regression for feature selection and missing value imputation, respectively, from a perspective that strengthens their performances. Further, different supervised machine learning models, the random forest (RF) model, support vector machine (SVM) model, and our designed twice-growth deep neural network (2GDNN) model are proposed for classification. The models are optimized by tuning the hyperparameters of the models using grid search and repeated stratified k-fold cross-validation and evaluated for their ability to scale to the prediction problem.

RESULTS

Through experiments on the PIMA Indian and LMCH diabetes datasets, precision, sensitivity, F1-score, train-accuracy, and test-accuracy scores of 97.34%, 97.24%, 97.26%, 99.01%, 97.25 and 97.28%, 97.33%, 97.27%, 99.57%, 97.33, are achieved with the proposed 2GDNN model, respectively.

CONCLUSION

The data preprocessing approaches and the classifiers with hyperparameter optimization proposed within the machine learning framework yield a robust machine learning model that outperforms state-of-the-art results in diabetes mellitus prediction and diagnosis. The source code for the models of the proposed machine learning framework has been made publicly available.

Higher levels of Bifidobacteria and tumor necrosis factor in children with drug-resistant epilepsy are associated with anti-seizure response to the ketogenic diet

Background

Recently, studies have suggested a role for the gut microbiota in epilepsy. Gut microbial changes during ketogenic diet (KD) treatment of drug-resistant epilepsy have been described. Inflammation is associated with certain types of epilepsy and specific inflammation markers decrease during KD. The gut microbiota plays an important role in the regulation of the immune system and inflammation.

Methods

28 children with drug-resistant epilepsy treated with the ketogenic diet were followed in this observational study. Fecal and serum samples were collected at baseline and three months after dietary intervention.

Findings

We identified both gut microbial and inflammatory changes during treatment. KD had a general anti-inflammatory effect. Novel bioinformatics and machine learning approaches identified signatures of specific Bifidobacteria and TNF (tumor necrosis factor) associated with responders before starting KD. During KD, taxonomic and inflammatory profiles between responders and non-responders were more similar than at baseline.

Interpretation

Our results suggest that children with drug-resistant epilepsy are more likely to benefit from KD treatment when specific Bifidobacteria and TNF are elevated. We here present a novel signature of interaction of the gut microbiota and the immune system associated with anti-epileptic response to KD treatment. This signature could be used as a prognostic biomarker to identify potential responders to KD before starting treatment. Our findings may also contribute to the development of new anti-seizure therapies by targeting specific components of the gut microbiota.

Funding

This study was supported by the Swedish Brain Foundation, Margarethahemmet Society, Stiftelsen Sunnerdahls Handikappfond, Linnea & Josef Carlssons Foundation, and The McCormick Genomic & Proteomic Center.

Urban Air Quality Assessment by Fusing Spatial and Temporal Data from Multiple Study Sources Using Refined Estimation Methods

In urban environmental management and public health evaluation efforts, there is an urgent need for fine-grained urban air quality monitoring. However, the high price and sparse distribution of air quality monitoring equipment make it difficult to develop effective and comprehensive fine-scale monitoring at the city scale. This has also led to air quality estimation methods based on incomplete monitoring data, which lack the ability to detect urban air quality differences within a neighborhood. To address this problem, this study proposes a refined urban air quality estimation method that fuses multisource spatio-temporal data. Based on the fact that urban air quality is easily affected by social activities, this method integrates meteorological data with urban social activity data to form a comprehensive environmental data set. It uses the spatio-temporal feature extraction model to extract the multi-source spatio-temporal features of the comprehensive environmental data set. Finally, the improved cascade forest algorithm is used to fit the relationship between the multisource spatio-temporal features and the air quality index (AQI) to construct an air quality estimation model, and the model is used to estimate the hourly PM2.5 index in Beijing on a 1 km × 1 km grid. The results show that the estimation model has excellent performance, and its goodness-of-fit (R2) and root mean square error (RMSE) reach 0.961 and 17.47, respectively. This method effectively achieves the assessment of urban air quality differences within a neighborhood and provides a new strategy for preventing information fragmentation and improving the effectiveness of information representation in the data fusion process.

Leveraging explainable machine learning to identify gait biomechanical parameters associated with anterior cruciate ligament injury

Anterior cruciate ligament (ACL) deficient and reconstructed knees display altered biomechanics during gait. Identifying significant gait changes is important for understanding normal and ACL function and is typically performed by statistical approaches. This paper focuses on the development of an explainable machine learning (ML) empowered methodology to: (i) identify important gait kinematic, kinetic parameters and quantify their contribution in the diagnosis of ACL injury and (ii) investigate the differences in sagittal plane kinematics and kinetics of the gait cycle between ACL deficient, ACL reconstructed and healthy individuals. For this aim, an extensive experimental setup was designed in which three-dimensional ground reaction forces and sagittal plane kinematic as well as kinetic parameters were collected from 151 subjects. The effectiveness of the proposed methodology was evaluated using a comparative analysis with eight well-known classifiers. Support Vector Machines were proved to be the best performing model (accuracy of 94.95%) on a group of 21 selected biomechanical parameters. Neural Networks accomplished the second best performance (92.89%). A state-of-the-art explainability analysis based on SHapley Additive exPlanations (SHAP) and conventional statistical analysis were then employed to quantify the contribution of the input biomechanical parameters in the diagnosis of ACL injury. Features, that would have been neglected by the traditional statistical analysis, were identified as contributing parameters having significant impact on the ML model’s output for ACL injury during gait.

Predicting Disease-Specific Survival for Patients With Primary Cholangiocarcinoma Undergoing Curative Resection by Using a Decision Tree Model

Background

The aim of this study was to derive and validate a decision tree model to predict disease-specific survival after curative resection for primary cholangiocarcinoma (CCA).

Method

Twenty-one clinical characteristics were collected from 482 patients after curative resection for primary CCA. A total of 289 patients were randomly allocated into a training cohort and 193 were randomly allocated into a validation cohort. We built three decision tree models based on 5, 12, and 21 variables, respectively. Area under curve (AUC), sensitivity, and specificity were used for comparison of the 0.5-, 1-, and 3-year decision tree models and regression models. AUC and decision curve analysis (DCA) were used to determine the predictive performances of the 0.5-, 1-, and 3-year decision tree models and AJCC TNM stage models.

Results

According to the fitting degree and the computational cost, the decision tree model derived from 12 variables displayed superior predictive efficacy to the other two models, with an accuracy of 0.938 in the training cohort and 0.751 in the validation cohort. Maximum tumor size, resection margin, lymph node status, histological differentiation, TB level, ALBI, AKP, AAPR, ALT, γ-GT, CA19-9, and Child-Pugh grade were involved in the model. The performances of 0.5-, 1-, and 3-year decision tree models were better than those of conventional models and AJCC TNM stage models.

Conclusion

We developed a decision tree model to predict outcomes for CCA undergoing curative resection. The present decision tree model outperformed other clinical models, facilitating individual decision-making of adjuvant therapy after curative resection.

Prediction of Field-Scale Wheat Yield Using Machine Learning Method and Multi-Spectral UAV Data

Accurate prediction of food crop yield is of great significance for global food security and regional trade stability. Since remote sensing data collected from unmanned aerial vehicle (UAV) platforms have the features of flexibility and high resolution, these data can be used as samples to develop regional regression models for accurate prediction of crop yield at a field scale. The primary objective of this study was to construct regional prediction models for winter wheat yield based on multi-spectral UAV data and machine learning methods. Six machine learning methods including Gaussian process regression (GPR), support vector machine regression (SVR) and random forest regression (RFR) were used for the construction of the yield prediction models. Ten vegetation indices (VIs) extracted from canopy spectral images of winter wheat acquired from a multi-spectral UAV at five key growth stages in Xuzhou City, Jiangsu Province, China in 2021 were selected as the variables of the models. In addition, in situ measurements of wheat yield were obtained in a destructive sampling manner for prediction algorithm modeling and validation. Prediction results of single growth stages showed that the optimal model was GPR constructed from extremely strong correlated VIs (ESCVIs) at the filling stage (R2 = 0.87, RMSE = 49.22 g/m2, MAE = 42.74 g/m2). The results of multiple stages showed GPR achieved the highest accuracy (R2 = 0.88, RMSE = 49.18 g/m2, MAE = 42.57 g/m2) when the ESCVIs of the flowering and filling stages were used. Larger sampling plots were adopted to verify the accuracy of yield prediction; the results indicated that the GPR model has strong adaptability at different scales. These findings suggest that using machine learning methods and multi-spectral UAV data can accurately predict crop yield at the field scale and deliver a valuable application reference for farm-scale field crop management.

Sports Medicine and Artificial Intelligence: A Primer

Artificial intelligence (AI) represents the fourth industrial revolution and the next frontier in medicine poised to transform the field of orthopaedics and sports medicine, though widespread understanding of the fundamental principles and adoption of applications remain nascent. Recent research efforts into implementation of AI in the field of orthopaedic surgery and sports medicine have demonstrated great promise in predicting athlete injury risk, interpreting advanced imaging, evaluating patient-reported outcomes, reporting value-based metrics, and augmenting the patient experience. Not unlike the recent emphasis thrust upon physicians to understand the business of medicine, the future practice of sports medicine specialists will require a fundamental working knowledge of the strengths, limitations, and applications of AI-based tools. With appreciation, caution, and experience applying AI to sports medicine, the potential to automate tasks and improve data-driven insights may be realized to fundamentally improve patient care. In this Current Concepts review, we discuss the definitions, strengths, limitations, and applications of AI from the current literature as it relates to orthopaedic sports medicine.

Prognostic Value of Right Ventricular Strains Using Novel Three-Dimensional Analytical Software in Patients With Cardiac Disease

Background

Right ventricular (RV) three-dimensional (3D) strains can be measured using novel 3D RV analytical software (ReVISION). Our objective was to investigate the prognostic value of RV 3D strains.

Methods

We retrospectively selected patients who underwent both 3D echocardiography (3DE) and cardiac magnetic resonance from January 2014 to October 2020. 3DE datasets were analyzed with 3D speckle tracking software and the ReVISION software. The primary end point was a composite of cardiac events, including cardiac death, heart failure hospitalization, or ventricular tachyarrhythmia.

Results

341 patients were included in this analysis. During a median of 20 months of follow-up, 49 patients reached a composite of cardiac events. In univariate analysis, 3D RV ejection fraction (RVEF) and three 3D strain values [RV global circumferential strain (3D RVGCS), RV global longitudinal strain (3D RVGLS), and RV global area strain (3D RVGAS)] were significantly associated with cardiac death, ventricular tachyarrhythmia, or heart failure hospitalization (Hazard ratio: 0.88 to 0.93, p < 0.05). Multivariate analysis revealed that 3D RVEF, three 3D strain values were significantly associated with cardiac events after adjusting for age, chronic kidney disease, and left ventricular systolic/diastolic parameters. Kaplan-Meier survival curves showed that 3D RVEF of 45% and median values of 3D RVGCS, 3D RVGLS, and 3D RVGAS stratified a higher risk for survival rates. Classification and regression tree analysis, including 22 clinical and echocardiographic parameters, selected 3D RVEF (cut-off value: 34.5%) first, followed by diastolic blood pressure (cut-off value: 53 mmHg) and 3D RVGAS (cut-off value: 32.4%) for stratifying two high-risk group, one intermediate-risk group, and one low-risk group.

Conclusions

RV 3D strain had an equivalent prognostic value compared with 3D RVEF. Combining these parameters with 3D RVEF may allow more detailed stratification of patient's prognosis in a wide array of cardiac diseases.

Optimizing size selectivity and catch patterns for hake (Merluccius merluccius) and blue whiting (Micromesistius poutassou) by combining square mesh panel and codend designs

Gear modifications in fisheries are usually implemented to obtain catch patterns that meet management objectives. In the Basque bottom trawl fishery, gear regulations include the use of a square mesh panel (SMP) placed at the top panel of the extension piece of the trawl to supplement diamond mesh codend selectivity. However, the catch patterns obtained with this combination have raised concern among scientists and authorities. This study combines new data on different SMP and codend designs with existing data from the literature to produce new results that are applied to predict the size selectivity and catch patterns of different gear combinations for a variety of fishing scenarios. A systematic approach based on the concept of treatment trees was outlined and applied to depict the effect of individual and combined gear design changes on size selectivity and catch patterns for hake (Merluccius merluccius) and blue whiting (Micromesistius poutassou). This approach led to identification of the gear combination with the most appropriate exploitation pattern for these two species and improved the readability and interpretation of selectivity results. The results demonstrated that changes both in SMP and, especially, codend designs have a significant effect on hake and blue whiting size selectivity and catch patterns. Therefore, we believe that further research should prioritize codend size selectivity, and additional selection devices may be added once codend designs with good selective properties are achieved.

Cognitive Status Predicts Return to Functional Independence After Minor Stroke: A Decision Tree Analysis

About two-thirds of patients with minor strokes are discharged home. However, these patients may have difficulties returning to their usual living activities. To investigate the factors associated with successful home discharge, our aim was to provide a decision tree (based on clinical data) that could identify if a patient discharged home could return to pre-stroke activities and to perform an external validation of this decision tree on an independent cohort. Two cohorts of patients with minor strokes gathered from stroke registries at the Hôpital Pitié-Salpêtrière and University Hospital Bern were included in this study (n = 105 for the construction cohort coming from France; n = 100 for the second cohort coming from Switzerland). The decision tree was built using the classification and regression tree (CART) analysis on the construction cohort. It was then applied to the validation cohort. Accuracy, sensitivity, specificity, false positive, and false-negative rates were reported for both cohorts. In the construction cohort, 60 patients (57%) returned to their usual, pre-stroke level of independence. The CART analysis produced a decision tree with the Montreal Cognitive Assessment (MoCA) as the first decision point, followed by discharge NIHSS score or age, and then by the occupational status. The overall prediction accuracy to the favorable outcome was 80% in the construction cohort and reached 72% accuracy in the validation cohort. This decision tree highlighted the role of cognitive function as a crucial factor for patients to return to their usual activities after a minor stroke. The algorithm may help clinicians to tailor planning of patients' discharge.

Machine Learning Techniques in Blood Pressure Management During the Acute Phase of Ischemic Stroke

Background and Purpose

Elevated blood pressure (BP) in acute ischemic stroke is common. A raised BP is related to mortality and disability, yet excessive BP lowering can be detrimental. The optimal BP management in acute ischemic stroke remains insufficient and relies on expert consensus statements. Permissive hypertension is recommended during the first 24-h after stroke onset, yet there is ongoing uncertainty regarding the most appropriate blood BP management in the acute phase of ischemic stroke. This study aims to develop a decision support tool for improving the management of extremely high BP during the first 24 h after acute ischemic stroke by using machine learning (ML) tools.

Methods

This diagnostic accuracy study used retrospective data from MIMIC-III and eICU databases. Decision trees were constructed by a hierarchical binary recursive partitioning algorithm to predict the BP-lowering of 10–30% off the maximal value when antihypertensive treatment was given in patients with an extremely high BP (above 220/110 or 180/105 mmHg for patients receiving thrombolysis), according to the American Heart Association/American Stroke Association (AHA/ASA), the European Society of Cardiology, and the European Society of Hypertension (ESC/ESH) guidelines. Regression trees were used to predict the time-weighted average BP. Implementation of synthetic minority oversampling technique was used to balance the dataset according to different antihypertensive treatments. The model performance of the decision tree was compared to the performance of neural networks, random forest, and logistic regression models.

Results

In total, 7,265 acute ischemic stroke patients were identified. Diastolic BP (DBP) is the main variable for predicting BP reduction in the first 24 h after a stroke. For patients receiving thrombolysis with DBP &lt;120 mmHg, Labetalol and Amlodipine are effective treatments. Above DBP of 120 mmHg, Amlodipine, Lisinopril, and Nicardipine are the most effective treatments. However, successful treatment depends on avoiding hyponatremia and on kidney functions.

Conclusion

This is the first study to address BP management in the acute phase of ischemic stroke using ML techniques. The results indicate that the treatment choice should be adjusted to different clinical and BP parameters, thus, providing a better decision-making approach.

A Deep Neural Network for Early Detection and Prediction of Chronic Kidney Disease

Diabetes and high blood pressure are the primary causes of Chronic Kidney Disease (CKD). Glomerular Filtration Rate (GFR) and kidney damage markers are used by researchers around the world to identify CKD as a condition that leads to reduced renal function over time. A person with CKD has a higher chance of dying young. Doctors face a difficult task in diagnosing the different diseases linked to CKD at an early stage in order to prevent the disease. This research presents a novel deep learning model for the early detection and prediction of CKD. This research objectives to create a deep neural network and compare its performance to that of other contemporary machine learning techniques. In tests, the average of the associated features was used to replace all missing values in the database. After that, the neural network's optimum parameters were fixed by establishing the parameters and running multiple trials. The foremost important features were selected by Recursive Feature Elimination (RFE). Hemoglobin, Specific Gravity, Serum Creatinine, Red Blood Cell Count, Albumin, Packed Cell Volume, and Hypertension were found as key features in the RFE. Selected features were passed to machine learning models for classification purposes. The proposed Deep neural model outperformed the other four classifiers (Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Logistic regression, Random Forest, and Naive Bayes classifier) by achieving 100% accuracy. The proposed approach could be a useful tool for nephrologists in detecting CKD.

Machine Learning-Based Model for Predicting Incidence and Severity of Acute Ischemic Stroke in Anterior Circulation Large Vessel Occlusion

Objectives: Patients with anterior circulation large vessel occlusion are at high risk of acute ischemic stroke, which could be disabling or fatal. In this study, we applied machine learning to develop and validate two prediction models for acute ischemic stroke (Model 1) and severity of neurological impairment (Model 2), both caused by anterior circulation large vessel occlusion (AC-LVO), based on medical history and neuroimaging data of patients on admission.

Methods: A total of 1,100 patients with AC- LVO from the Second Hospital of Hebei Medical University in North China were enrolled, of which 713 patients presented with acute ischemic stroke (AIS) related to AC- LVO and 387 presented with the non-acute ischemic cerebrovascular event. Among patients with the non-acute ischemic cerebrovascular events, 173 with prior stroke or TIA were excluded. Finally, 927 patients with AC-LVO were entered into the derivation cohort. In the external validation cohort, 150 patients with AC-LVO from the Hebei Province People's Hospital, including 99 patients with AIS related to AC- LVO and 51 asymptomatic AC-LVO patients, were retrospectively reviewed. We developed four machine learning models [logistic regression (LR), regularized LR (RLR), support vector machine (SVM), and random forest (RF)], whose performance was internally validated using 5-fold cross-validation. The performance of each machine learning model for the area under the receiver operating characteristic curve (ROC-AUC) was compared and the variables of each algorithm were ranked.

Results: In model 1, among the included patients with AC-LVO, 713 (76.9%) and 99 (66%) suffered an acute ischemic stroke in the derivation and external validation cohorts, respectively. The ROC-AUC of LR, RLR and SVM were significantly higher than that of the RF in the external validation cohorts [0.66 (95% CI 0.57–0.74) for LR, 0.66 (95% CI 0.57–0.74) for RLR, 0.55 (95% CI 0.45–0.64) for RF and 0.67 (95% CI 0.58–0.76) for SVM]. In model 2, 254 (53.9%) and 31 (37.8%) patients suffered disabling ischemic stroke in the derivation and external validation cohorts, respectively. There was no difference in AUC among the four machine learning algorithms in the external validation cohorts.

Conclusions: Machine learning methods with multiple clinical variables have the ability to predict acute ischemic stroke and the severity of neurological impairment in patients with AC-LVO.