Applications of artificial neural networks in health care organizational decision-making: A scoping review

Label-free rapid antimicrobial susceptibility testing with machine-learning based dynamic holographic laser speckle imaging

Antimicrobial resistance (AMR) presents a significant global challenge, creating an urgent need for rapid and sensitive antimicrobial susceptibility testing (AST) methods to guide timely treatment decisions. Traditional AST techniques, such as broth microdilution, disk diffusion, and gradient diffusion assays, require extended incubation times, delaying critical therapeutic interventions. In this study, we present a dynamic holographic laser speckle imaging (DhLSI) system, coupled with machine learning algorithms, for rapid assessment of bacterial susceptibility upon antibiotic treatment. Our method operates by utilizing a reference beam to enhance the detection of weak scattering signals, capable of performing AST at bacterial concentrations as low as 103 CFU/mL, while producing results consistent with those obtained using the standard concentration of 105 CFU/mL. By employing artificial neural networks (ANN) to analyze dynamic speckle patterns, the DhLSI system can determine bacterial susceptibility within 2-3 h. The system was validated using model Gram-positive and Gram-negative bacterial strains, as well as two antibiotic treatments with different mechanisms of action. Experiments conducted on bacteria incubated on different days demonstrated consistent performance. This approach offers a rapid, label-free platform for early-stage infection diagnosis and effective antimicrobial stewardship, with the potential to be implemented in clinical settings to address AMR challenges.

Applications of machine learning and deep learning in musculoskeletal medicine: a narrative review

Artificial intelligence (AI), with its technologies such as machine perception, robotics, natural language processing, expert systems, and machine learning (ML) with its subset deep learning, have transformed patient care and administration in all fields of modern medicine. For many clinicians, however, the nature, scope, and resulting possibilities of ML and deep learning might not yet be fully clear. This narrative review provides an overview of the application of ML and deep learning in musculoskeletal medicine. It first introduces the concept of AI and machine learning and its associated fields. Different machine concepts such as supervised, unsupervised and reinforcement learning will then be presented with current applications and clinical perspective. Finally deep learning applications will be discussed. With significant improvements over the last decade, ML and its subset deep learning today offer potent tools for numerous applications to implement in clinical practice. While initial setup costs are high, these investments can reduce workload and cost globally. At the same time, many challenges remain, such as standardisation in data labelling and often insufficient validity of the obtained results. In addition, legal aspects still will have to be clarified. Until good analyses and predictions are obtained by an ML tool, patience in training and suitable data sets are required. Awareness of the strengths of ML and the limitations that lie within it will help put this technique to good use.

Probabilistic design space exploration and optimization via bayesian approach for a fluid bed drying process

The concept of Design Space (DS), delineated as a region of investigated variables aimed at maintaining product quality, was introduced in the International Conference on Harmonisation (ICH) Q8 as a framework to direct pharmaceutical development. However, the complexity of processes and the presence of uncertainties in pharmaceutical manufacturing exacerbate the difficulties of exploring a reliable and robust DS. This study investigates the probabilistic design space to explain the process operability and performance reliability using a Bayesian approach for a fluid bed drying process. We initially develop a Bayesian model by integrating a surrogate-based predictive model with embedded uncertainty quantification of material variability. Subsequently, employing a grid search-based technique to discretize the operational variable domain facilitates the exploration of the probabilistic DS to meet the specified product quality requirements. Meanwhile, optimization is employed to obtain the maximum DS region and enhance its operability. Results demonstrate that the Bayesian approach is an effective method to identify a probability DS to guarantee product quality at the desired reliability level considering material and process uncertainty.

The impact of AI-driven sentiment analysis on patient outcomes in psychiatric care: A narrative review

This article addresses the pressing question of how advanced analytical tools, specifically artificial intelligence (AI)-driven sentiment analysis, can be effectively integrated into psychiatric care to enhance patient outcomes. Utilizing specific search phrases like "AI-driven sentiment analysis," "psychiatric care," and "patient outcomes," a comprehensive survey of English-language publications from the years 2014-2024 was performed. This examination encompassed multiple databases such as PubMed, PsycINFO, Google Scholar, and IEEE Xplore. Through a comprehensive analysis of qualitative case studies and quantitative metrics, the study uncovered that the implementation of sentiment analysis significantly improves clinicians' ability to monitor and respond to patient emotions, leading to more tailored treatment plans and increased patient engagement. Key findings indicated that sentiment analysis improves early mood disorder detection, personalizes treatments, enhances patient-provider communication, and boosts treatment adherence, leading to better mental health outcomes. The significance of these findings lies in their potential to revolutionize psychiatric care by providing healthcare professionals with real-time insights into patient feelings and responses, thereby facilitating more proactive and empathetic care strategies. Furthermore, this study highlights the broader implications for healthcare systems, suggesting that the incorporation of sentiment analysis can lead to a paradigm shift in how mental health services are delivered, ultimately enhancing the efficacy and quality of care. By addressing barriers to new technology adoption and demonstrating its practical benefits, this research contributes vital knowledge to the ongoing discourse on optimizing healthcare delivery through innovative solutions in psychiatric settings.

Machine Learning-Based Prediction Models for Healthcare Outcomes in Patients Participating in Cardiac Rehabilitation: A Systematic Review

PURPOSE

Cardiac rehabilitation (CR) has been proven to reduce mortality and morbidity in patients with cardiovascular disease. Machine learning (ML) techniques are increasingly used to predict healthcare outcomes in various fields of medicine including CR. This systemic review aims to perform critical appraisal of existing ML-based prognosis predictive model within CR and identify key research gaps in this area.

REVIEW METHODS

A systematic literature search was conducted in Scopus, PubMed, Web of Science, and Google Scholar from the inception of each database to January 28, 2024. The data extracted included clinical features, predicted outcomes, model development, and validation as well as model performance metrics. Included studies underwent quality assessments using the IJMEDI and Prediction Model Risk of Bias Assessment Tool checklist.

SUMMARY

A total of 22 ML-based clinical models from 7 studies across multiple phases of CR were included. Most models were developed using smaller patient cohorts from 41 to 227, with one exception involving 2280 patients. The prediction objectives ranged from patient intention to initiate CR to graduate from outpatient CR along with interval physiological and psychological progression in CR. The best-performing ML models reported area under the receiver operating characteristics curve between 0.82 and 0.91, with sensitivity from 0.77 to 0.95, indicating good prediction capabilities. However, none of them underwent calibration or external validation. Most studies raised concerns about bias. Readiness of these models for implementation into practice is questionable. External validation of existing models and development of new models with robust methodology based on larger populations and targeting diverse clinical outcomes in CR are needed.

Application of an Innovative Methodology to Build Infrastructure for Digital Transformation of Health Systems: Developmental Program Evaluation

BACKGROUND

The current public health crises we face, including communicable disease pandemics such as COVID-19, require cohesive societal efforts to address decision-making gaps in our health systems. Digital health platforms that leverage big data ethically from citizens can transform health systems by enabling real-time data collection, communication, and rapid responses. However, the lack of standardized and evidence-based methods to develop and implement digital health platforms currently limits their application.

OBJECTIVE

This study aims to apply mixed evaluation methods to assess the development of a rapid response COVID-19 digital health platform before public launch by engaging with the development and research team, which consists of interdisciplinary researchers (ie, key stakeholders).

METHODS

Using a developmental evaluation approach, this study conducted (1) a qualitative survey assessing digital health platform objectives, modifications, and challenges administered to 5 key members of the software development team and (2) a role-play pilot with 7 key stakeholders who simulated 8 real-world users, followed by a self-report survey, to evaluate the utility of the digital health platform for each of its objectives. Survey data were analyzed using an inductive thematic analysis approach. Postpilot test survey data were aggregated and synthesized by participant role.

RESULTS

The digital health platform met original objectives and was expanded to accommodate the evolving needs of potential users and COVID-19 pandemic regulations. Key challenges noted by the development team included navigating changing government policies and supporting the data sovereignty of platform users. Strong team cohesion and problem-solving were essential in the overall success of program development. During the pilot test, participants reported positive experiences interacting with the platform and found its features relatively easy to use. Users in the community member role felt that the platform accurately reflected their risk of contracting COVID-19, but reported some challenges interacting with the interface. Those in the decision maker role found the data visualizations helpful for understanding complex information. Both participant groups highlighted the utility of a tutorial for future users.

CONCLUSIONS

Evaluation of the digital health platform development process informed our decisions to integrate the research team more cohesively with the development team, a practice that is currently uncommon given the use of external technology vendors in health research. In the short term, the developmental evaluation resulted in shorter sprints, and the role-play exercise enabled improvements to the log-in process and user interface ahead of public deployment. In the long term, this exercise informed the decision to include a data scientist as part of both teams going forward to liaise with researchers throughout the development process. More interdisciplinarity was also integrated into the research process by providing health system training to computer programmers, a key factor in human-centered artificial intelligence development.

Navigating digital frontiers in UAE healthcare: A qualitative exploration of healthcare professionals' and patients' experiences with AI and telemedicine

The integration of artificial intelligence (AI) and telemedicine into healthcare has significantly advanced patient-centered care, enhancing accessibility, convenience, and patient-doctor relationships. However, different factors determine the extent to which such benefits are realized, especially in unique healthcare settings such as the United Arab Emirates (UAE). In this regard, this research explores healthcare professionals' and patients' perspectives to understand various factors that influence the adoption and use of AI in the UAE's healthcare sector. This research sought to understand the benefits, challenges, and enablers of successful adoption and utilization of AI and telemedicine in the UAE's healthcare setting. Through this objective, this research aims to contribute to the scanty knowledge on the integration of emerging technologies, such as AI, in different infrastructural and cultural contexts. The study employed a qualitative approach, through which eight healthcare professionals and seven patients (totaling 15 participants) were recruited from Dubai- and Abu Dhabi-based hospitals using the purposive sampling strategy. The participants' insights and views on the research topic were captured using semi-structured face-to-face interviews. These interviews were analyzed using the thematic analysis strategy. This study established that while AI and telemedicine are associated with various benefits, including enhancing the management of chronic illnesses, effective controlling of infectious diseases, saving patients and hospitals health-related costs and time, and enhancing convenience, they suffer from various drawbacks, including limited infrastructural and financial resources, significant gaps in skills, safety concerns, and the likelihood of misdiagnosis and misinformation. The study also observed that the successful integration of AI and telemedicine in the UAE healthcare sector necessitated incentivizing stakeholders to use this technology, full involvement and engagement of stakeholders across all stages of implementation, adequate training of the healthcare staff, and public engagement and awareness. This research demonstrates that integrating AI and telemedicine in the UAE healthcare sector necessitates addressing contextual infrastructural and cultural hindrances. The results highlight the need to address such limitations, adequately train healthcare professionals, and enhance data privacy. The study also lays a foundation for further research into contextual challenges hindering the effective adoption and implementation of AI and telemedicine in different healthcare settings in order to develop a generic, context-specific framework that will guide the adoption of such emerging technologies in the global healthcare industry.

Prediction of malnutrition in kids by integrating ResNet-50-based deep learning technique using facial images

In recent times, severe acute malnutrition (SAM) in India is considered a serious issue as per UNICEF 2022 records. In that record, 35.5% of children under age 5 are stunted, 19.3% are wasted, and 32% are underweight. Malnutrition, defined as these three conditions, affects 5.7 million children globally. This research utilizes an artificial intelligence-based image segmentation technique to predict malnutrition in children. The primary goal of this research is to use a deep learning model to eliminate the need for multiple manual diagnostic tests and simplify the prediction of malnutrition in kids. The traditional model uses text-based data and takes more time with continuous monitoring of kids by analysing body mass index (BMI) over different periods. Children in rural areas often miss medical expert appointments, and a lack of knowledge among parents can lead to severe malnutrition. The aim of the proposed system is to eliminate the need for manual blood tests and regular visits to medical experts. This study uses the ResNet-50 deep learning model's built-in shortcut connection to solve the image-based vanishing gradient problem. This makes training more efficient for image segmentation tasks in predicting malnutrition. The model is 98.49% accurate in predicting the kids who are malnourished among the kids who are healthy. It is evident from the results that the proposed system serves better than other deep learning models, such as XG Boost (75.29% accuracy), VGG 16 (94% accuracy), Xception (95.41% accuracy), and MobileNet (92.42% accuracy). Hence, the proposed technique is effective in detecting malnutrition and diagnose it earlier, without using predictive analysis function or advice from the medical experts.

Artificial Intelligence and Predictive Modeling in the Management and Treatment of Episodic Migraine

PURPOSE OF REVIEW

Artificial intelligence (AI) has impacted different aspects of headache medicine, from history taking and diagnosis to drug development. AI has been shown to have predictive modeling in helping diagnose migraine and assist with patient care. Additionally, this technology has been adapted to help non-headache specialists with headache management. Similar practices have expanded to help diagnose cluster headache. AI has also been used to help streamline patient visits, and identify new drug targets.

RECENT FINDINGS

Various forms of AI models have been implemented in headache medicine; these have ranged from diagnosis engines to models helping track headache triggers. Additionally, AI has been used to assist in clinical trials and to help predict placebo responses to different medications. There are still several limitations with AI in setting of headache medicine. AI and diagnosis models have a role to play in headache medicine. However, technology is still in its infancy and limitations do exist.

Developing decision making framework on built-up site suitability assessment for urban regeneration in the industrial cities of Eastern India

Unprecedented urban growth in developing countries impacts the existing urban planning as well as prospective urban regeneration. Therefore, evaluating the prospective suitable sites for built-up area development is important to make sustainable urban planning through the urban regeneration process in the industrial-based urban area Asansol Municipal Corporation (AMC). Hence, we analyzed the area-specific built-up suitability using machine learning soft-computing techniques: Artificial Neural Network, Random Forest, and Support Vector Machine. The result showed that the edge of the urban center and periphery of the Asansol, Kulti, and Raniganj were found to be very high (21.52%, 19.87%, 26.32%) to high suitable (11.48%, 19%, 27.26%) areas for further urban planning due to vacant land with available services nearby. However, the southern portion, especially along the Damodar River site and the area near the mining sites were found to be low to very low suitable zones due to inadequate service facilities and high pollution. Finally, we proposed a three-tier urban regeneration framework for sustainable built-up development strategies in AMC that helps to achieve the UN's sustainable development goals-3, 8, 11, 12 and 13. The findings of this study will benefit policymakers by pointing out the ideal areas for suitable built-up area development initiatives in the near future.

Artificial intelligence applied to bed regulation in Rio Grande do Norte: Data analysis and application of machine learning on the "RegulaRN Leitos Gerais" platform

Bed regulation within Brazil's National Health System (SUS) plays a crucial role in managing care for patients in need of hospitalization. In Rio Grande do Norte, Brazil, the RegulaRN Leitos Gerais platform was the information system developed to register requests for bed regulation for COVID-19 cases. However, the platform was expanded to cover a range of diseases that require hospitalization. This study explored different machine learning models in the RegulaRN database, from October 2021 to January 2024, totaling 47,056 regulations. From the data obtained, 12 features were selected from the 24 available. After that, blank and inconclusive data were removed, as well as the outcomes that had values other than discharge and death, rendering a binary classification. Data was also correlated, balanced, and divided into training and test portions for application in machine learning models. The results showed better accuracy (87.77%) and recall (87.77%) for the XGBoost model, and higher precision (87.85%) and F1-Score (87.56%) for the Random Forest and Gradient Boosting models, respectively. As for Specificity (82.94%) and ROC-AUC (82.13%), the Multilayer Perceptron with SGD optimizer obtained the highest scores. The results evidenced which models could adequately assist medical regulators during the decision-making process for bed regulation, enabling even more effective regulation and, consequently, greater availability of beds and a decrease in waiting time for patients.

Predicting maternal risk level using machine learning models

BACKGROUND

Maternal morbidity and mortality remain critical health concerns globally. As a result, reducing the maternal mortality ratio (MMR) is part of goal 3 in the global sustainable development goals (SDGs), and previously, it was an important indicator in the Millennium Development Goals (MDGs). Therefore, identifying high-risk groups during pregnancy is crucial for decision-makers and medical practitioners to mitigate mortality and morbidity. However, the availability of accurate predictive models for maternal mortality and maternal health risks is challenging. Compared with traditional predictive models, machine learning algorithms have emerged as promising predictive modelling methods providing accurate predictive models.

METHODS

This work aims to explore the potential of machine learning (ML) algorithms in maternal risk level prediction using a nationwide maternal mortality dataset from Oman for the first time. A total of 402 maternal deaths from 1991 to 2023 in Oman were included in this study. We utilised principal component analysis (PCA) in the ML algorithms and compared them to the results of model performance without PCA. We employed and compared ten ML algorithms, including decision tree (DT), random forest (RF), K-Nearest Neighbors (KNN), Naïve Bayes (NB), Extreme Gradient Boosting (xgboost), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Logistic Regression (LR), Support Vector Machine (SVM) and Artificial Neural Network (ANN). Different metrics, including, accuracy, sensitivity, precision, and the F1- score, were utilised to assess Model performance.

RESULTS

The results indicated that the RF model outperformed the other methods in predicting the risk level (low or high) with an accuracy of 75.2%, precision of 85.7% and F1- score of 73% after PCA was applied.

CONCLUSIONS

We applied several machine learning models to predict maternal risk levels for the first time using real data from Oman. RF outperformed the other algorithms in this classification problem. A reliable estimate of maternal risk level would facilitate intervention plans for medical practitioners to reduce maternal death.

Enhancing Autism Detection Through Gaze Analysis Using Eye Tracking Sensors and Data Attribution with Distillation in Deep Neural Networks

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by differences in social communication and repetitive behaviors, often associated with atypical visual attention patterns. In this paper, the Gaze-Based Autism Classifier (GBAC) is proposed, which is a Deep Neural Network model that leverages both data distillation and data attribution techniques to enhance ASD classification accuracy and explainability. Using data sampled by eye tracking sensors, the model identifies unique gaze behaviors linked to ASD and applies an explainability technique called TracIn for data attribution by computing self-influence scores to filter out noisy or anomalous training samples. This refinement process significantly improves both accuracy and computational efficiency, achieving a test accuracy of 94.35% while using only 77% of the dataset, showing that the proposed GBAC outperforms the same model trained on the full dataset and random sample reductions, as well as the benchmarks. Additionally, the data attribution analysis provides insights into the most influential training examples, offering a deeper understanding of how gaze patterns correlate with ASD-specific characteristics. These results underscore the potential of integrating explainable artificial intelligence into neurodevelopmental disorder diagnostics, advancing clinical research by providing deeper insights into the visual attention patterns associated with ASD.

Optimization of fungal secondary metabolites production via response surface methodology coupled with multi-parameter optimized artificial neural network model

Filamentous fungi's secondary metabolites (SMs) possess significant application owing to their distinct structure and diverse bioactivities, yet their restricted yield levels often hinder further research and application. The study developed a response surface methodology-artificial neural network (RSM-ANN) strategy with multi-parameter optimizations of the ANN model to optimize medium for the production of two high-value fungal SMs, echinocandin E and paraherquamide A. Multi-parameter optimization of the ANN model was achieved through stratifying experimental data, fully adjusting neural network internals, and evaluating metaheuristic algorithms for optimal initial weights and biases. Experimental validation of models revealed that ANN-genetic algorithm models outperformed traditional RSM models in terms of determination coefficients, accuracy, and mean squared errors. ANN models showed outstanding robustness across a variety of fungal species, mediums, and experimental designs (Central Composite Design or Box-Behnken Design). This work refines the RSM-ANN optimization technique to increase fungal SM production efficiency, enabling industrial-scale production and applications.

Artificial Intelligence in Surgery: A Systematic Review of Use and Validation

Background: Artificial Intelligence (AI) holds promise for transforming healthcare, with AI models gaining increasing clinical use in surgery. However, new AI models are developed without established standards for their validation and use. Before AI can be widely adopted, it is crucial to ensure these models are both accurate and safe for patients. Without proper validation, there is a risk of integrating AI models into practice without sufficient evidence of their safety and accuracy, potentially leading to suboptimal patient outcomes. In this work, we review the current use and validation methods of AI models in clinical surgical settings and propose a novel classification system. Methods: A systematic review was conducted in PubMed and Cochrane using the keywords "validation", "artificial intelligence", and "surgery", following PRISMA guidelines. Results: The search yielded a total of 7627 articles, of which 102 were included for data extraction, encompassing 2,837,211 patients. A validation classification system named Surgical Validation Score (SURVAS) was developed. The primary applications of models were risk assessment and decision-making in the preoperative setting. Validation methods were ranked as high evidence in only 45% of studies, and only 14% of the studies provided publicly available datasets. Conclusions: AI has significant applications in surgery, but validation quality remains suboptimal, and public data availability is limited. Current AI applications are mainly focused on preoperative risk assessment and are suggested to improve decision-making. Classification systems such as SURVAS can help clinicians confirm the degree of validity of AI models before their application in practice.

Prevention of adverse HIV treatment outcomes: machine learning to enable proactive support of people at risk of HIV care disengagement in Tanzania

OBJECTIVES

This study aimed to develop a machine learning (ML) model to predict disengagement from HIV care, high viral load or death among people living with HIV (PLHIV) with the goal of enabling proactive support interventions in Tanzania. The algorithm addressed common challenges when applying ML to electronic medical record (EMR) data: (1) imbalanced outcome distribution; (2) heterogeneity across multisite EMR data and (3) evolving virological suppression thresholds.

DESIGN

Observational study using a national EMR database.

SETTING

Conducted in two regions in Tanzania, using data from the National HIV Care database.

PARTICIPANTS

The study included over 6 million HIV care visit records from 295 961 PLHIV in two regions in Tanzania's National HIV Care database from January 2015 to May 2023.

RESULTS

Our ML model effectively identified PLHIV at increased risk of adverse outcomes. Key predictors included past disengagement from care, antiretroviral therapy (ART) status (which tracks a patient's engagement with ART across visits), age and time on ART. The downsampling approach we implemented effectively managed imbalanced data to reduce prediction bias. Site-specific algorithms performed better compared with a universal approach, highlighting the importance of tailoring ML models to local contexts. A sensitivity analysis confirmed the model's robustness to changes in viral load suppression thresholds.

CONCLUSIONS

ML models leveraging large-scale databases of patient data offer significant potential to identify PLHIV for interventions to enhance engagement in HIV care in resource-limited settings. Tailoring algorithms to local contexts and flexibility towards evolving clinical guidelines are essential for maximising their impact.

Machine Learning versus Cox Models for Predicting Overall Survival in Patients with Osteosarcoma: A Retrospective Analysis of the EURAMOS-1 Clinical Trial Data

Since the mid-1980s, there has been little progress in improving survival of patients diagnosed with osteosarcoma. Survival prediction models play a key role in clinical decision-making, guiding healthcare professionals in tailoring treatment strategies based on individual patient risks. The increasing interest of the medical community in using machine learning (ML) for predicting survival has sparked an ongoing debate on the value of ML techniques versus more traditional statistical modelling (SM) approaches. This study investigates the use of SM versus ML methods in predicting overall survival (OS) using osteosarcoma data from the EURAMOS-1 clinical trial (NCT00134030). The well-established Cox proportional hazard model is compared to the extended Cox model that includes time-varying effects, and to the ML methods random survival forests and survival neural networks. The impact of eight variables on OS predictions is explored. Results are compared on different model performance metrics, variable importance, and patient-specific predictions. The article provides comprehensive insights to aid healthcare researchers in evaluating diverse survival prediction models for low-dimensional clinical data.

Self-learning activation functions to increase accuracy of privacy-preserving Convolutional Neural Networks with homomorphic encryption

The widespread adoption of cloud computing necessitates privacy-preserving techniques that allow information to be processed without disclosure. This paper proposes a method to increase the accuracy and performance of privacy-preserving Convolutional Neural Networks with Homomorphic Encryption (CNN-HE) by Self-Learning Activation Functions (SLAF). SLAFs are polynomials with trainable coefficients updated during training, together with synaptic weights, for each polynomial independently to learn task-specific and CNN-specific features. We theoretically prove its feasibility to approximate any continuous activation function to the desired error as a function of the SLAF degree. Two CNN-HE models are proposed: CNN-HE-SLAF and CNN-HE-SLAF-R. In the first model, all activation functions are replaced by SLAFs, and CNN is trained to find weights and coefficients. In the second one, CNN is trained with the original activation, then weights are fixed, activation is substituted by SLAF, and CNN is shortly re-trained to adapt SLAF coefficients. We show that such self-learning can achieve the same accuracy 99.38% as a non-polynomial ReLU over non-homomorphic CNNs and lead to an increase in accuracy (99.21%) and higher performance (6.26 times faster) than the state-of-the-art CNN-HE CryptoNets on the MNIST optical character recognition benchmark dataset.

Prognosis of COVID-19 severity using DERGA, a novel machine learning algorithm

It is important to determine the risk for admission to the intensive care unit (ICU) in patients with COVID-19 presenting at the emergency department. Using artificial neural networks, we propose a new Data Ensemble Refinement Greedy Algorithm (DERGA) based on 15 easily accessible hematological indices. A database of 1596 patients with COVID-19 was used; it was divided into 1257 training datasets (80 % of the database) for training the algorithms and 339 testing datasets (20 % of the database) to check the reliability of the algorithms. The optimal combination of hematological indicators that gives the best prediction consists of only four hematological indicators as follows: neutrophil-to-lymphocyte ratio (NLR), lactate dehydrogenase, ferritin, and albumin. The best prediction corresponds to a particularly high accuracy of 97.12 %. In conclusion, our novel approach provides a robust model based only on basic hematological parameters for predicting the risk for ICU admission and optimize COVID-19 patient management in the clinical practice.

Prediction equations for detecting COVID-19 infection using basic laboratory parameters

Objectives

Coronavirus disease 2019 (COVID-19) emerged as a global pandemic during 2019 to 2022. The gold standard method of detecting this disease is reverse transcription-polymerase chain reaction (RT-PCR). However, RT-PCR has a number of shortcomings. Hence, the objective is to propose a cheap and effective method of detecting COVID-19 infection by using machine learning (ML) techniques, which encompasses five basic parameters as an alternative to the costly RT-PCR.

Materials and Methods

Two machine learning-based predictive models, namely, Artificial Neural Network (ANN) and Multivariate Adaptive Regression Splines (MARS), are designed for predicting COVID-19 infection as a cheaper and simpler alternative to RT-PCR utilizing five basic parameters [i.e., age, total leucocyte count, red blood cell count, platelet count, C-reactive protein (CRP)]. Each of these parameters was studied, and correlation is drawn with COVID-19 diagnosis and progression. These laboratory parameters were evaluated in 171 patients who presented with symptoms suspicious of COVID-19 in a hospital at Kharagpur, India, from April to August 2022. Out of a total of 171 patients, 88 and 83 were found to be COVID-19-negative and COVID-19-positive, respectively.

Results

The accuracies of the predicted class are found to be 97.06% and 91.18% for ANN and MARS, respectively. CRP is found to be the most significant input parameter. Finally, two predictive mathematical equations for each ML model are provided, which can be quite useful to detect the COVID-19 infection easily.

Conclusion

It is expected that the present study will be useful to the medical practitioners for predicting the COVID-19 infection in patients based on only five very basic parameters.

The Utility of Artificial Intelligence and Machine Learning in the Diagnosis of Takotsubo Cardiomyopathy: A Systematic Review

Introduction:

Takotsubo cardiomyopathy (TTC) is a cardiovascular disease caused by physical/psychological stressors with significant morbidity if left untreated. Because TTC often mimics acute myocardial infarction in the absence of obstructive coronary disease, the condition is often underdiagnosed in the population. Our aim was to discuss the role of artificial intelligence (AI) and machine learning (ML) in diagnosing TTC.

Methods:

We systematically searched electronic databases from inception until April 8, 2023, for studies on the utility of AI- or ML-based algorithms in diagnosing TTC compared with other cardiovascular diseases or healthy controls. We summarized major findings in a narrative fashion and tabulated relevant numerical parameters.

Results:

Five studies with a total of 920 patients were included. Four hundred and forty-seven were diagnosed with TTC via International Classification of Diseases codes or the Mayo Clinic diagnostic criteria, while there were 473 patients in the comparator group (29 of healthy controls, 429 of myocardial infarction, and 14 of acute myocarditis). Hypertension and smoking were the most common comorbidities in both cohorts, but there were no statistical differences between TTC and comparators. Two studies utilized deep-learning algorithms on transthoracic echocardiographic images, while the rest incorporated supervised ML on cardiac magnetic resonance imaging, 12-lead electrocardiographs, and brain magnetic resonance imaging. All studies found that AI-based algorithms can increase the diagnostic rate of TTC when compared to healthy controls or myocardial infarction patients. In three of these studies, AI-based algorithms had higher sensitivity and specificity compared to human readers.

Conclusion:

AI and ML algorithms can improve the diagnostic capacity of TTC and additionally reduce erroneous human error in differentiating from MI and healthy individuals.

A theory of oral healthcare decision-making in Appalachia

INTRODUCTION

People make oral healthcare decisions regardless of having partial information, misinformation, sources that deliberately mislead, or information that is culturally influenced. This is particularly true in the Appalachian culture where oral healthcare decision-making practices are not well understood by researchers and dental professionals. Despite efforts to improve dental care utilization, the Appalachia region remains low in oral healthcare utilization. There is a need for a theory to identify concepts in decision-making when seeking oral healthcare. The theory could be useful in creating oral health interventions. The study objective is to develop a theory to identify concepts that influence oral healthcare decision-making in Appalachia (OHDA).

METHODS

The researchers used a grounded theory qualitative study design to explain data for a theory of OHDA. Participants from Appalachia, in 20-minute interviews, provided insights into concepts that influence OHDA from August 22, 2017 to May 26, 2022. Notes/memos were written during and after the interviews and coding was conducted after the interviews. Open coding categories emerged through constant comparison of responses.

RESULTS

Five overarching concepts that embody OHDA were discovered: Affect (Level of Pain/Emotion/Stress involvement), Awareness, Trust/belief, Resources, and Risk Perception. All participants discussed the impact of social media toward these concepts.

CONCLUSION

To influence a person's OHDA, public health officials and researchers need to address the person's affect, level of awareness, trust/belief, available resources, and risk perception. Social media is very important in awareness concerning oral health information. These factors are important to consider for similar research in oral healthcare utilization at the population level.

Predicting Male Infertility Using Artificial Neural Networks: A Review of the Literature

Male infertility is a relevant public health problem, but there is no systematic review of the different machine learning (ML) models and their accuracy so far. The present review aims to comprehensively investigate the use of ML algorithms in predicting male infertility, thus reporting the accuracy of the used models in the prediction of male infertility as a primary outcome. Particular attention will be paid to the use of artificial neural networks (ANNs). A comprehensive literature search was conducted in PubMed, Scopus, and Science Direct between 15 July and 23 October 2023, conducted under the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We performed a quality assessment of the included studies using the recommended tools suggested for the type of study design adopted. We also made a screening of the Risk of Bias (RoB) associated with the included studies. Thus, 43 relevant publications were included in this review, for a total of 40 different ML models detected. The studies included reported a good quality, even if RoB was not always good for all the types of studies. The included studies reported a median accuracy of 88% in predicting male infertility using ML models. We found only seven studies using ANN models for male infertility prediction, reporting a median accuracy of 84%.

Comparison of ANN and LR models for predicting Carbapenem-resistant Klebsiella pneumoniae isolates from a southern province of China's RNSS data

OBJECTIVES

Carbapenem-resistant Klebsiella pneumoniae (CRKP) is a serious threat to public health due to its limited treatment options and high mortality rate. This study aims to identify the risk factors of carbapenem resistance in patients with K. pneumoniae isolates and develop CRKP prediction models using logistic regression (LR) and artificial neural network (ANN) methods.

METHODS

We retrospectively analysed the data of 49,774 patients with Klebsiella pneumoniae isolates from a regional nosocomial infection surveillance system (RNSS) between 2018 and 2021. We performed logistic regression analyses to determine the independent predictors for CRKP. We then built and evaluated LR and ANN models based on these predictors using calibration curves, ROC curves, and decision curve analysis (DCA). We also applied the Synthetic Minority Over-Sampling Technique (SMOTE) to balance the data of CRKP and non-CRKP groups.

RESULTS

The LR model showed good discrimination and calibration in both training and validation sets, with areas under the ROC curve (AUROC) of 0.824 and 0.825, respectively. The DCA indicated that the LR model had clinical usefulness for decision making. The ANN model outperformed the LR model both in the training set and validation set. The SMOTE technique improved the performance of both models for CRKP detection in training set, but not in the validation set.

CONCLUSION

We developed and validated LR and ANN models for predicting CRKP based on RNSS data. Both models were feasible and reliable for CRKP inference and could potentially assist clinicians in selecting appropriate empirical antibiotics and reducing unnecessary medical resource utilization.

Developing an intelligent prediction system for successful aging based on artificial neural networks

Background

Due to the growing number of disabilities in elderly, Attention to this period of life is essential to be considered. Few studies focused on the physical, mental, disabilities, and disorders affecting the quality of life in elderly people. SA1 is related to various factors influencing the elderly's life. So, the objective of the current study is to build an intelligent system for SA prediction through ANN2 algorithms to investigate better all factors affecting the elderly life and promote them.

Methods

This study was performed on 1156 SA and non-SA cases. We applied statistical feature reduction method to obtain the best factors predicting the SA. Two models of ANNs with 5, 10, 15, and 20 neurons in hidden layers were used for model construction. Finally, the best ANN configuration was obtained for predicting the SA using sensitivity, specificity, accuracy, and cross-entropy loss function.

Results

The study showed that 25 factors correlated with SA at the statistical level of P < 0.05. Assessing all ANN structures resulted in FF-BP3 algorithm having the configuration of 25-15-1 with accuracy-train of 0.92, accuracy-test of 0.86, and accuracy-validation of 0.87 gaining the best performance over other ANN algorithms.

Conclusions

Developing the CDSS for predicting SA has crucial role to effectively inform geriatrics and health care policymakers decision making.

Navigating the Future of Cardiac Diagnostics: Insights From Artificial Neural Networks

Cardiovascular diseases remain a leading cause of mortality globally, necessitating innovative approaches for early detection and precise diagnostic methodologies. Artificial neural networks (ANNs), inspired by the complexity of the human brain's neural networks, have emerged as powerful tools for transforming the landscape of cardiac diagnostics. ANNs are capable of learning complex patterns from data. In cardiac diagnostics, these networks are employed to analyze intricate cardiovascular data, providing insights into diseases such as coronary artery disease and arrhythmias. From personalized medicine approaches to predictive analytics, ANNs can revolutionize the identification of cardiovascular risks, enabling timely interventions and preventive measures. The integration of ANNs with wearable devices and telemedicine is poised to establish a connected healthcare ecosystem, providing holistic and continuous cardiac monitoring. However, challenges persist, including ethical considerations surrounding patient data and uncertainties in diagnostic outcomes. Looking forward, the prospects of ANNs in cardiac diagnostics are promising. Anticipated technological advancements and collaborative efforts between medical and technological communities are expected to drive innovation, address current challenges, and foster a new era of precision cardiac care.

A cluster-based ensemble approach for congenital heart disease prediction

BACKGROUND

One of the most prevalent birth disorders is congenital heart diseases (CHD). Although CHD risk factors have been the subject of numerous studies, their propensity to cause CHD has not been tested. Particularly few research has attempted to forecast CHD risk using population-based cross-sectional data, which is inherently imbalanced.

OBJECTIVE

The main goals of this study are to create a reliable data analysis model that can help with (i) a better understanding of congenital heart disease prediction in the presence of missing and unbalanced data and (ii) creating cohorts of expectant mothers with similar lifestyle characteristics.

METHODS

Clusters of patient cohorts are produced using the unsupervised data mining technique density-based spatial clustering of applications with noise (DBSCAN). For more accurate CHD prediction, a random forest model was trained using these clusters and their corresponding patterns. This study uses a dataset of 33,831 expectant mothers to make its prediction. Missing data were handled using the k-NN imputation approach, while extremely unbalanced data were balanced using SMOTE. These techniques are all data-driven and need little to no user or expert involvement.

RESULTS AND CONCLUSION

Using DBSCAN, three cohorts were found. The cluster information enhanced the random forest-based CHD prediction and revealed intricate factors that influence prediction accuracy. The proposed approach gave the highest results with 99 % accuracy and 0.91 AUC and performed better than the state-of-the-art methodologies. Hence, the suggested method using unsupervised learning can provide intricate information to the classifier and further enhance the performance of the classification.

Machine learning and topological data analysis identify unique features of human papillae in 3D scans

The tongue surface houses a range of papillae that are integral to the mechanics and chemistry of taste and textural sensation. Although gustatory function of papillae is well investigated, the uniqueness of papillae within and across individuals remains elusive. Here, we present the first machine learning framework on 3D microscopic scans of human papillae ([Formula: see text]), uncovering the uniqueness of geometric and topological features of papillae. The finer differences in shapes of papillae are investigated computationally based on a number of features derived from discrete differential geometry and computational topology. Interpretable machine learning techniques show that persistent homology features of the papillae shape are the most effective in predicting the biological variables. Models trained on these features with small volumes of data samples predict the type of papillae with an accuracy of 85%. The papillae type classification models can map the spatial arrangement of filiform and fungiform papillae on a surface. Remarkably, the papillae are found to be distinctive across individuals and an individual can be identified with an accuracy of 48% among the 15 participants from a single papillae. Collectively, this is the first evidence demonstrating that tongue papillae can serve as a unique identifier, and inspires a new research direction for food preferences and oral diagnostics.

Fostering Excellence in Knee Arthroplasty: Developing Optimal Patient Care Pathways and Inspiring Knowledge Transfer of Advanced Surgical Techniques

Osteoarthritis of the knee is common. Early sports trauma or cartilage defects are risk factors for osteoarthritis. If conservative treatment fails, partial or total joint replacement is often performed. A joint replacement aims to restore physiological biomechanics and the quality of life of affected patients. Total knee arthroplasty is one of the most performed surgeries in musculoskeletal medicine. Several developments have taken place over the last decades that have truly altered the way we look at knee arthroplasty today. Some of the fascinating aspects will be presented and discussed in the present narrative review.

Predicting Risk of Cardiac Arrest in Young Asian Americans: Insights from an Artificial Neural Network Analysis of the Nationwide Cohort

We used the Artificial Neural Network (ANN) model to identify predictors of Sudden Cardiac Arrest (SCA) in a national cohort of young Asian patients in the United States. The National Inpatient Sample (2019) was used to identify young Asians (18-44-year-old) who were hospitalized with SCA. The neural network's predicted criteria for SCA were selected. After eliminating missing data, young Asians (n = 65,413) were randomly divided into training (n = 45,094) and testing (n = 19347) groups. Training data (70%) was used to calibrate ANN while testing data (30%) was utilized to assess the algorithm's accuracy. To determine ANN's performance in predicting SCA, we compared the frequency of incorrect prediction between training and testing data and measured the area under the Receiver Operating Curve (AUC). The 2019 young Asian cohort had 327,065 admissions (median age 32 years; 84.2% female), with SCA accounting for 0.21%. The exact rate of error in predictions vs. tests was shown by training data (0.2% vs 0.2%). In descending order, the normalized importance of predictors to accurately predict SCA in young adults included prior history of cardiac arrest, sex, age, diabetes, anxiety disorders, prior coronary artery bypass grafting, hypertension, congenital heart disease, income, peripheral vascular disease, and cancer. The AUC was 0.821, indicating an excellent ANN model for SCA prediction. Our ANN models performed excellently in revealing the order of important predictors of SCA in young Asian American patients. These findings could have a considerable impact on clinical practice to develop risk prediction models to improve the survival outcome in high-risk patients.

Wearable-based human flow experience recognition enhanced by transfer learning methods using emotion data

BACKGROUND

Flow experience is a specific positive and affective state that occurs when humans are completely absorbed in an activity and forget everything else. This state can lead to high performance, well-being, and productivity at work. Few studies have been conducted to determine the human flow experience using physiological wearable sensor devices. Other studies rely on self-reported data.

METHODS

In this article, we use physiological data collected from 25 subjects with multimodal sensing devices, in particular the Empatica E4 wristband, the Emotiv Epoc X electroencephalography (EEG) headset, and the Biosignalplux RespiBAN - in arithmetic and reading tasks to automatically discriminate between flow and non-flow states using feature engineering and deep feature learning approaches. The most meaningful wearable device for flow detection is determined by comparing the performances of each device. We also investigate the connection between emotions and flow by testing transfer learning techniques involving an emotion recognition-related task on the source domain.

RESULTS

The EEG sensor modalities yielded the best performances with an accuracy of 64.97%, and a macro Averaged F1 (AF1) score of 64.95%. An accuracy of 73.63% and an AF1 score of 72.70% were obtained after fusing all sensor modalities from all devices. Additionally, our proposed transfer learning approach using emotional arousal classification on the DEAP dataset led to an increase in performances with an accuracy of 75.10% and an AF1 score of 74.92%.

CONCLUSION

The results of this study suggest that effective discrimination between flow and non-flow states is possible with multimodal sensor data. The success of transfer learning using the DEAP emotion dataset as a source domain indicates that emotions and flow are connected, and emotion recognition can be used as a latent task to enhance the performance of flow recognition.

Machine learning risk estimation and prediction of death in continuing care facilities using administrative data

In this study, we aimed to identify the factors that were associated with mortality among continuing care residents in Alberta, during the coronavirus disease 2019 (COVID-19) pandemic. We achieved this by leveraging and linking various administrative datasets together. Then, we examined pre-processing methods in terms of prediction performance. Finally, we developed several machine learning models and compared the results of these models in terms of performance. We conducted a retrospective cohort study of all continuing care residents in Alberta, Canada, from March 1, 2020, to March 31, 2021. We used a univariable and a multivariable logistic regression (LR) model to identify predictive factors of 60-day all-cause mortality by estimating odds ratios (ORs) with a 95% confidence interval. To determine the best sensitivity-specificity cut-off point, the Youden index was employed. We developed several machine learning models to determine the best model regarding performance. In this cohort study, increased age, male sex, symptoms, previous admissions, and some specific comorbidities were associated with increased mortality. Machine learning and pre-processing approaches offer a potentially valuable method for improving risk prediction for mortality, but more work is needed to show improvement beyond standard risk factors.

Exploring the Steps of Infrared (IR) Spectral Analysis: Pre-Processing, (Classical) Data Modelling, and Deep Learning

Infrared (IR) spectroscopy has greatly improved the ability to study biomedical samples because IR spectroscopy measures how molecules interact with infrared light, providing a measurement of the vibrational states of the molecules. Therefore, the resulting IR spectrum provides a unique vibrational fingerprint of the sample. This characteristic makes IR spectroscopy an invaluable and versatile technology for detecting a wide variety of chemicals and is widely used in biological, chemical, and medical scenarios. These include, but are not limited to, micro-organism identification, clinical diagnosis, and explosive detection. However, IR spectroscopy is susceptible to various interfering factors such as scattering, reflection, and interference, which manifest themselves as baseline, band distortion, and intensity changes in the measured IR spectra. Combined with the absorption information of the molecules of interest, these interferences prevent direct data interpretation based on the Beer-Lambert law. Instead, more advanced data analysis approaches, particularly artificial intelligence (AI)-based algorithms, are required to remove the interfering contributions and, more importantly, to translate the spectral signals into high-level biological/chemical information. This leads to the tasks of spectral pre-processing and data modeling, the main topics of this review. In particular, we will discuss recent developments in both tasks from the perspectives of classical machine learning and deep learning.

Predicting the drop out from the maternal, newborn and child healthcare continuum in three East African Community countries: application of machine learning models

BACKGROUND

For optimal health, the maternal, newborn, and child healthcare (MNCH) continuum necessitates that the mother/child receive the full package of antenatal, intrapartum, and postnatal care. In sub-Saharan Africa, dropping out from the MNCH continuum remains a challenge. Using machine learning, the study sought to forecast the MNCH continuum drop out and determine important predictors in three East African Community (EAC) countries.

METHODS

The study utilised Demographic Health Surveys data from the Democratic Republic of Congo (DRC) (2013/14), Kenya (2014) and Tanzania (2015/16). STATA 17 was used to perform the multivariate logistic regression. Python 3.0 was used to build five machine learning classification models namely the Logistic Regression, Random Forest, Decision Tree, Support Vector Machine and Artificial Neural Network. Performance of the models was assessed using Accuracy, Precision, Recall, Specificity, F1 score and area under the Receiver Operating Characteristics (AUROC).

RESULTS

The prevalence of the drop out from the MNCH continuum was 91.0% in the DRC, 72.4% in Kenya and 93.6% in Tanzania. Living in the rural areas significantly increased the odds of dropping out from the MNCH continuum in the DRC (AOR:1.76;95%CI:1.30-2.38), Kenya (AOR:1.23;95%CI:1.03-1.47) and Tanzania (AOR:1.41;95%CI:1.01-1.97). Lower maternal education also conferred a significant increase in the DRC (AOR:2.16;95%CI:1.67-2.79), Kenya (AOR:1.56;95%CI:1.30-1.84) and Tanzania (AOR:1.70;95%CI:1.24-2.34). Non exposure to mass media also conferred a significant positive influence in the DRC (AOR:1.49;95%CI:1.15-1.95), Kenya (AOR:1.46;95%CI:1.19-1.80) and Tanzania (AOR:1.65;95%CI:1.13-2.40). The Random Forest exhibited superior predictive accuracy (Accuracy = 75.7%, Precision = 79.1%, Recall = 92.1%, Specificity = 51.6%, F1 score = 85.1%, AUROC = 70%). The top four predictors with the greatest influence were household wealth, place of residence, maternal education and exposure to mass media.

CONCLUSIONS

The MNCH continuum dropout rate is very high in the EAC countries. Maternal education, place of residence, and mass media exposure were common contributing factors to the drop out from MNCH continuum. The Random Forest had the highest predictive accuracy. Household wealth, place of residence, maternal education and exposure to mass media were ranked among the top four features with significant influence. The findings of this study can be used to support evidence-based decisions in MNCH interventions and to develop web-based services to improve continuity of care retention.

Revolutionising dental technologies: a qualitative study on dental technicians' perceptions of Artificial intelligence integration

BACKGROUND

The integration of artificial intelligence (AI) in dentistry has the potential to revolutionise the field of dental technologies. However, dental technicians' views on the use of AI in dental technology are still sparse in the literature. This qualitative study aimed to explore the perceptions of dental technicians regarding the use of AI in their dental laboratory practice.

METHODS

Twelve dental technicians with at least five years of professional experience and currently working in Malaysia agreed to participate in the one-to-one in-depth online interviews. Interviews were recorded, transcribed verbatim and translated. Thematic analysis was conducted to identify patterns, themes, and categories within the interview transcripts.

RESULTS

The analysis revealed two key themes: "Perceived Benefits of AI" and "Concerns and Challenges". Dental technicians recognised the enhanced efficiency, productivity, accuracy, and precision that AI can bring to dental laboratories. They also acknowledged the streamlined workflow and improved communication facilitated by AI systems. However, concerns were raised regarding job security, professional identity, ethical considerations, and the need for adequate training and support.

CONCLUSION

This research sheds light on the potential benefits and challenges associated with the integration of AI in dental laboratory practices. Understanding these perceptions and addressing the challenges can support the effective integration of AI in dental laboratories and contribute to the growing body of literature on AI in healthcare.

Applications of Natural Language Processing for the Management of Stroke Disorders: Scoping Review

BACKGROUND

Recent advances in natural language processing (NLP) have heightened the interest of the medical community in its application to health care in general, in particular to stroke, a medical emergency of great impact. In this rapidly evolving context, it is necessary to learn and understand the experience already accumulated by the medical and scientific community.

OBJECTIVE

The aim of this scoping review was to explore the studies conducted in the last 10 years using NLP to assist the management of stroke emergencies so as to gain insight on the state of the art, its main contexts of application, and the software tools that are used.

METHODS

Data were extracted from Scopus and Medline through PubMed, using the keywords "natural language processing" and "stroke." Primary research questions were related to the phases, contexts, and types of textual data used in the studies. Secondary research questions were related to the numerical and statistical methods and the software used to process the data. The extracted data were structured in tables and their relative frequencies were calculated. The relationships between categories were analyzed through multiple correspondence analysis.

RESULTS

Twenty-nine papers were included in the review, with the majority being cohort studies of ischemic stroke published in the last 2 years. The majority of papers focused on the use of NLP to assist in the diagnostic phase, followed by the outcome prognosis, using text data from diagnostic reports and in many cases annotations on medical images. The most frequent approach was based on general machine learning techniques applied to the results of relatively simple NLP methods with the support of ontologies and standard vocabularies. Although smaller in number, there has been an increasing body of studies using deep learning techniques on numerical and vectorized representations of the texts obtained with more sophisticated NLP tools.

CONCLUSIONS

Studies focused on NLP applied to stroke show specific trends that can be compared to the more general application of artificial intelligence to stroke. The purpose of using NLP is often to improve processes in a clinical context rather than to assist in the rehabilitation process. The state of the art in NLP is represented by deep learning architectures, among which Bidirectional Encoder Representations from Transformers has been found to be especially widely used in the medical field in general, and for stroke in particular, with an increasing focus on the processing of annotations on medical images.

Using the Field Artificial Intelligence Triage (FAIT) tool to predict hospital critical care resource utilization in patients with truncal gunshot wounds

BACKGROUND

Tiered trauma triage systems have resulted in a significant mortality reduction, but models have remained unchanged. The aim of this study was to develop and test an artificial intelligence algorithm to predict critical care resource utilization.

METHODS

We queried the ACS-TQIP 2017-18 database for truncal gunshot wounds(GSW). An information-aware deep neural network (DNN-IAD) model was trained to predict ICU admission and need for mechanical ventilation (MV). Input variables included demographics, comorbidities, vital signs, and external injuries. The model's performance was assessed using the area under receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC).

RESULTS

For the ICU admission analysis, we included 39,916 patients. For the MV need analysis, 39,591 patients were included. Median (IQR) age was 27 (22,36). AUROC and AUPRC for predicting ICU need were 84.8 ± 0.5 and 75.4 ± 0.5, and the AUROC and AUPRC for MV need were 86.8 ± 0.5 and 72.5 ± 0.6.

CONCLUSIONS

Our model predicts hospital utilization outcomes in patients with truncal GSW with high accuracy, allowing early resource mobilization and rapid triage decisions in hospitals with capacity issues and austere environments.

Secure medical image transmission using deep neural network in e-health applications

Recently, medical technologies have developed, and the diagnosis of diseases through medical images has become very important. Medical images often pass through the branches of the network from one end to the other. Hence, high-level security is required. Problems arise due to unauthorized use of data in the image. One of the methods used to secure data in the image is encryption, which is one of the most effective techniques in this field. Confusion and diffusion are the two main steps addressed here. The contribution here is the adaptation of the deep neural network by the weight that has the highest impact on the output, whether it is an intermediate output or a semi-final output in additional to a chaotic system that is not detectable using deep neural network algorithm. The colour and grayscale images were used in the proposed method by dividing the images according to the Region of Interest by the deep neural network algorithm. The algorithm was then used to generate random numbers to randomly create a chaotic system based on the replacement of columns and rows, and randomly distribute the pixels on the designated area. The proposed algorithm evaluated in several ways, and compared with the existing methods to prove the worth of the proposed method.

An intelligent decision support approach for quantified assessment of innovation ability via an improved BP neural network

In today's competitive and changing social environment, innovation and entrepreneurial ability have become important factors for the successful development of college students. However, relying solely on traditional evaluation methods and indicators cannot comprehensively and accurately evaluate the innovation and entrepreneurial potential and ability of college students. Therefore, developing a comprehensive evaluation model is urgently needed. To address this issue, this article introduces machine learning methods to explore the learning ability of subjective evaluation processes and proposes an intelligent decision support method for quantitatively evaluating innovation capabilities using an improved BP (Back Propagation) neural network. This article first introduces the current research status of evaluating the innovation and entrepreneurship ability of college students, and based on previous research, it has been found that inconsistent evaluation standards are one of the important issues at present. Then, based on different BP models and combined with the actual situation of college student innovation and entrepreneurship evaluation, we selected an appropriate input layer setting for the BP neural network and improved the setting of the middle layer (hidden layer). The identification of output nodes was also optimized by combining the current situation. Subsequently, the conversion function, initial value and threshold were determined. Finally, evaluation indicators were determined and an improved BP model was established which was validated using examples. The research results indicate that the improved BP neural network model has a low error rate, strong generalization ability and ideal prediction effect which can be effectively used to analyze problems related to intelligent evaluation of innovation ability.

Deep hybrid model for maternal health risk classification in pregnancy: synergy of ANN and random forest

Introduction

Maternal health is a critical aspect of public health that affects the wellbeing of both mothers and infants. Despite medical advancements, maternal mortality rates remain high, particularly in developing countries. AI-based models provide new ways to analyze and interpret medical data, which can ultimately improve maternal and fetal health outcomes.

Methods

This study proposes a deep hybrid model for maternal health risk classification in pregnancy, which utilizes the strengths of artificial neural networks (ANN) and random forest (RF) algorithms. The proposed model combines the two algorithms to improve the accuracy and efficiency of risk classification in pregnant women. The dataset used in this study consists of features such as age, systolic and diastolic blood pressure, blood sugar, body temperature, and heart rate. The dataset is divided into training and testing sets, with 75% of the data used for training and 25% used for testing. The output of the ANN and RF classifier is considered, and a maximum probability voting system selects the output with the highest probability as the most correct.

Results

Performance is evaluated using various metrics, such as accuracy, precision, recall, and F1 score. Results showed that the proposed model achieves 95% accuracy, 97% precision, 97% recall, and an F1 score of 0.97 on the testing dataset.

Discussion

The deep hybrid model proposed in this study has the potential to improve the accuracy and efficiency of maternal health risk classification in pregnancy, leading to better health outcomes for pregnant women and their babies. Future research could explore the generalizability of this model to other populations, incorporate unstructured medical data, and evaluate its feasibility for clinical use.

Use of machine learning to predict bladder cancer survival outcomes: a systematic literature review

INTRODUCTION

The objective of this systematic review is to summarize the use of machine learning (ML) in predicting overall survival (OS) in patients with bladder cancer.

METHODS

Search terms for bladder cancer, ML algorithms, and mortality were used to identify studies in PubMed and Web of Science as of February 2022. Notable inclusion/exclusion criteria contained the inclusion of studies that utilized patient-level datasets and exclusion of primary gene expression-related dataset studies. Study quality and bias were assessed using the International Journal of Medical Informatics (IJMEDI) checklist.

RESULTS

Of the 14 included studies, the most common algorithms were artificial neural networks (n = 8) and logistic regression (n = 4). Nine articles described missing data handling, with five articles removing patients with missing data entirely. With respect to feature selection, the most common sociodemographic variables were age (n = 9), gender (n = 9), and smoking status (n = 3), with clinical variables most commonly including tumor stage (n = 8), grade (n = 7), and lymph node involvement (n = 6). Most studies (n = 10) were of medium IJMEDI quality, with common areas of improvement being the descriptions of data preparation and deployment.

CONCLUSIONS

ML holds promise for optimizing bladder cancer care through accurate OS predictions, but challenges related to data processing, feature selection, and data source quality must be resolved to develop robust models. While this review is limited by its inability to compare models across studies, this systematic review will inform decision-making by various stakeholders to improve understanding of ML-based OS prediction in bladder cancer and foster interpretability of future models.

A multiphysics-based artificial neural networks model for atherosclerosis

Atherosclerosis is a medical condition involving the hardening and/or thickening of arteries' walls. Mathematical multi-physics models have been developed to predict the development of atherosclerosis under different conditions. However, these models are typically computationally expensive. In this study, we used machine learning techniques, particularly artificial neural networks (ANN), to enhance the computational efficiency of these models. A database of multi-physics Finite Element Method (FEM) simulations was created and used for training and validating an ANN model. The model is capable of quick and accurate prediction of atherosclerosis development. A remarkable computational gain is obtained using the ANN model compared to the original FEM simulations.

Digital Health Dashboards for Decision-Making to Enable Rapid Responses During Public Health Crises: Replicable and Scalable Methodology

BACKGROUND

The COVID-19 pandemic has reiterated the need for cohesive, collective, and deliberate societal efforts to address inherent inefficiencies in our health systems and overcome decision-making gaps using real-time data analytics. To achieve this, decision makers need independent and secure digital health platforms that engage citizens ethically to obtain big data, analyze and convert big data into real-time evidence, and finally, visualize this evidence to inform rapid decision-making.

OBJECTIVE

The objective of this study is to develop replicable and scalable jurisdiction-specific digital health dashboards for rapid decision-making to ethically monitor, mitigate, and manage public health crises via systems integration beyond health care.

METHODS

The primary approach in the development of the digital health dashboard was the use of global digital citizen science to tackle pandemics like COVID-19. The first step in the development process was to establish an 8-member Citizen Scientist Advisory Council via Digital Epidemiology and Population Health Laboratory's community partnerships. Based on the consultation with the council, three critical needs of citizens were prioritized: (1) management of household risk of COVID-19, (2) facilitation of food security, and (3) understanding citizen accessibility of public services. Thereafter, a progressive web application (PWA) was developed to provide daily services that address these needs. The big data generated from citizen access to these PWA services are set up to be anonymized, aggregated, and linked to the digital health dashboard for decision-making, that is, the dashboard displays anonymized and aggregated data obtained from citizen devices via the PWA. The digital health dashboard and the PWA are hosted on the Amazon Elastic Compute Cloud server. The digital health dashboard's interactive statistical navigation was designed using the Microsoft Power Business Intelligence tool, which creates a secure connection with the Amazon Relational Database server to regularly update the visualization of jurisdiction-specific, anonymized, and aggregated data.

RESULTS

The development process resulted in a replicable and scalable digital health dashboard for decision-making. The big data relayed to the dashboard in real time reflect usage of the PWA that provides households the ability to manage their risk of COVID-19, request food when in need, and report difficulties and issues in accessing public services. The dashboard also provides (1) delegated community alert system to manage risks in real time, (2) bidirectional engagement system that allows decision makers to respond to citizen queries, and (3) delegated access that provides enhanced dashboard security.

CONCLUSIONS

Digital health dashboards for decision-making can transform public health policy by prioritizing the needs of citizens as well as decision makers to enable rapid decision-making. Digital health dashboards provide decision makers the ability to directly communicate with citizens to mitigate and manage existing and emerging public health crises, a paradigm-changing approach, that is, inverting innovation by prioritizing community needs, and advancing digital health for equity.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

RR1-10.2196/46810.

Improvement of predictive accuracies of functional outcomes after subacute stroke inpatient rehabilitation by machine learning models

OBJECTIVES

Stepwise linear regression (SLR) is the most common approach to predicting activities of daily living at discharge with the Functional Independence Measure (FIM) in stroke patients, but noisy nonlinear clinical data decrease the predictive accuracies of SLR. Machine learning is gaining attention in the medical field for such nonlinear data. Previous studies reported that machine learning models, regression tree (RT), ensemble learning (EL), artificial neural networks (ANNs), support vector regression (SVR), and Gaussian process regression (GPR), are robust to such data and increase predictive accuracies. This study aimed to compare the predictive accuracies of SLR and these machine learning models for FIM scores in stroke patients.

METHODS

Subacute stroke patients (N = 1,046) who underwent inpatient rehabilitation participated in this study. Only patients' background characteristics and FIM scores at admission were used to build each predictive model of SLR, RT, EL, ANN, SVR, and GPR with 10-fold cross-validation. The coefficient of determination (R2) and root mean square error (RMSE) values were compared between the actual and predicted discharge FIM scores and FIM gain.

RESULTS

Machine learning models (R2 of RT = 0.75, EL = 0.78, ANN = 0.81, SVR = 0.80, GPR = 0.81) outperformed SLR (0.70) to predict discharge FIM motor scores. The predictive accuracies of machine learning methods for FIM total gain (R2 of RT = 0.48, EL = 0.51, ANN = 0.50, SVR = 0.51, GPR = 0.54) were also better than of SLR (0.22).

CONCLUSIONS

This study suggested that the machine learning models outperformed SLR for predicting FIM prognosis. The machine learning models used only patients' background characteristics and FIM scores at admission and more accurately predicted FIM gain than previous studies. ANN, SVR, and GPR outperformed RT and EL. GPR could have the best predictive accuracy for FIM prognosis.

MIDO GDM: an innovative artificial intelligence-based prediction model for the development of gestational diabetes in Mexican women

Given the barriers to early detection of gestational diabetes mellitus (GDM), this study aimed to develop an artificial intelligence (AI)-based prediction model for GDM in pregnant Mexican women. Data were retrieved from 1709 pregnant women who participated in the multicenter prospective cohort study 'Cuido mi embarazo'. A machine-learning-driven method was used to select the best predictive variables for GDM risk: age, family history of type 2 diabetes, previous diagnosis of hypertension, pregestational body mass index, gestational week, parity, birth weight of last child, and random capillary glucose. An artificial neural network approach was then used to build the model, which achieved a high level of accuracy (70.3%) and sensitivity (83.3%) for identifying women at high risk of developing GDM. This AI-based model will be applied throughout Mexico to improve the timing and quality of GDM interventions. Given the ease of obtaining the model variables, this model is expected to be clinically strategic, allowing prioritization of preventative treatment and promising a paradigm shift in prevention and primary healthcare during pregnancy. This AI model uses variables that are easily collected to identify pregnant women at risk of developing GDM with a high level of accuracy and precision.

Statistical models versus machine learning for competing risks: development and validation of prognostic models

BACKGROUND

In health research, several chronic diseases are susceptible to competing risks (CRs). Initially, statistical models (SM) were developed to estimate the cumulative incidence of an event in the presence of CRs. As recently there is a growing interest in applying machine learning (ML) for clinical prediction, these techniques have also been extended to model CRs but literature is limited. Here, our aim is to investigate the potential role of ML versus SM for CRs within non-complex data (small/medium sample size, low dimensional setting).

METHODS

A dataset with 3826 retrospectively collected patients with extremity soft-tissue sarcoma (eSTS) and nine predictors is used to evaluate model-predictive performance in terms of discrimination and calibration. Two SM (cause-specific Cox, Fine-Gray) and three ML techniques are compared for CRs in a simple clinical setting. ML models include an original partial logistic artificial neural network for CRs (PLANNCR original), a PLANNCR with novel specifications in terms of architecture (PLANNCR extended), and a random survival forest for CRs (RSFCR). The clinical endpoint is the time in years between surgery and disease progression (event of interest) or death (competing event). Time points of interest are 2, 5, and 10 years.

RESULTS

Based on the original eSTS data, 100 bootstrapped training datasets are drawn. Performance of the final models is assessed on validation data (left out samples) by employing as measures the Brier score and the Area Under the Curve (AUC) with CRs. Miscalibration (absolute accuracy error) is also estimated. Results show that the ML models are able to reach a comparable performance versus the SM at 2, 5, and 10 years regarding both Brier score and AUC (95% confidence intervals overlapped). However, the SM are frequently better calibrated.

CONCLUSIONS

Overall, ML techniques are less practical as they require substantial implementation time (data preprocessing, hyperparameter tuning, computational intensity), whereas regression methods can perform well without the additional workload of model training. As such, for non-complex real life survival data, these techniques should only be applied complementary to SM as exploratory tools of model's performance. More attention to model calibration is urgently needed.

Risk factor identification and prediction models for prolonged length of stay in hospital after acute ischemic stroke using artificial neural networks

Background

Accurate estimation of prolonged length of hospital stay after acute ischemic stroke provides crucial information on medical expenditure and subsequent disposition. This study used artificial neural networks to identify risk factors and build prediction models for a prolonged length of stay based on parameters at the time of hospitalization.

Methods

We retrieved the medical records of patients who received acute ischemic stroke diagnoses and were treated at a stroke center between January 2016 and June 2020, and a retrospective analysis of these data was performed. Prolonged length of stay was defined as a hospital stay longer than the median number of days. We applied artificial neural networks to derive prediction models using parameters associated with the length of stay that was collected at admission, and a sensitivity analysis was performed to assess the effect of each predictor. We applied 5-fold cross-validation and used the validation set to evaluate the classification performance of the artificial neural network models.

Results

Overall, 2,240 patients were enrolled in this study. The median length of hospital stay was 9 days. A total of 1,101 patients (49.2%) had a prolonged hospital stay. A prolonged length of stay is associated with worse neurological outcomes at discharge. Univariate analysis identified 14 baseline parameters associated with prolonged length of stay, and with these parameters as input, the artificial neural network model achieved training and validation areas under the curve of 0.808 and 0.788, respectively. The mean accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of prediction models were 74.5, 74.9, 74.2, 75.2, and 73.9%, respectively. The key factors associated with prolonged length of stay were National Institutes of Health Stroke Scale scores at admission, atrial fibrillation, receiving thrombolytic therapy, history of hypertension, diabetes, and previous stroke.

Conclusion

The artificial neural network model achieved adequate discriminative power for predicting prolonged length of stay after acute ischemic stroke and identified crucial factors associated with a prolonged hospital stay. The proposed model can assist in clinically assessing the risk of prolonged hospitalization, informing decision-making, and developing individualized medical care plans for patients with acute ischemic stroke.

Estimating the Depth of Anesthesia from EEG Signals Based on a Deep Residual Shrinkage Network

The reliable monitoring of the depth of anesthesia (DoA) is essential to control the anesthesia procedure. Electroencephalography (EEG) has been widely used to estimate DoA since EEG could reflect the effect of anesthetic drugs on the central nervous system (CNS). In this study, we propose that a deep learning model consisting mainly of a deep residual shrinkage network (DRSN) and a 1 × 1 convolution network could estimate DoA in terms of patient state index (PSI) values. First, we preprocessed the four raw channels of EEG signals to remove electrical noise and other physiological signals. The proposed model then takes the preprocessed EEG signals as inputs to predict PSI values. Then we extracted 14 features from the preprocessed EEG signals and implemented three conventional feature-based models as comparisons. A dataset of 18 patients was used to evaluate the models' performances. The results of the five-fold cross-validation show that there is a relatively high similarity between the ground-truth PSI values and the predicted PSI values of our proposed model, which outperforms the conventional models, and further, that the Spearman's rank correlation coefficient is 0.9344. In addition, an ablation experiment was conducted to demonstrate the effectiveness of the soft-thresholding module for EEG-signal processing, and a cross-subject validation was implemented to illustrate the robustness of the proposed method. In summary, the procedure is not merely feasible for estimating DoA by mimicking PSI values but also inspired us to develop a precise DoA-estimation system with more convincing assessments of anesthetization levels.

Application of artificial intelligence in medical technologies: A systematic review of main trends

Objective

Artificial intelligence (AI) has been increasingly applied in various fields of science and technology. In line with the current research, medicine involves an increasing number of artificial intelligence technologies. The introduction of rapid AI can lead to positive and negative effects. This is a multilateral analytical literature review aimed at identifying the main branches and trends in the use of using artificial intelligence in medical technologies.

Methods

The total number of literature sources reviewed is n = 89, and they are analyzed based on the literature reporting evidence-based guideline PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) for a systematic review.

Results

As a result, from the initially selected 198 references, 155 references were obtained from the databases and the remaining 43 sources were found on open internet as direct links to publications. Finally, 89 literature sources were evaluated after exclusion of unsuitable references based on the duplicated and generalized information without focusing on the users.

Conclusions

This article is identifying the current state of artificial intelligence in medicine and prospects for future use. The findings of this review will be useful for healthcare and AI professionals for improving the circulation and use of medical AI from design to implementation stage.

Development of a peripheral blood transcriptomic gene signature to predict bronchopulmonary dysplasia

Bronchopulmonary dysplasia (BPD) is the most common lung disease of extreme prematurity, yet mechanisms that associate with or identify neonates with increased susceptibility for BPD are largely unknown. Combining artificial intelligence with gene expression data is a novel approach that may assist in better understanding mechanisms underpinning chronic lung disease and in stratifying patients at greater risk for BPD. The objective of this study is to develop an early peripheral blood transcriptomic signature that can predict preterm neonates at risk for developing BPD. Secondary analysis of whole blood microarray data from 97 very low birth weight neonates on day of life 5 was performed. BPD was defined as positive pressure ventilation or oxygen requirement at 28 days of age. Participants were randomly assigned to a training (70%) and testing cohort (30%). Four gene-centric machine learning models were built, and their discriminatory abilities were compared with gestational age or birth weight. This study adheres to the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) statement. Neonates with BPD (n = 62 subjects) exhibited a lower median gestational age (26.0 wk vs. 30.0 wk, P < 0.01) and birth weight (800 g vs. 1,280 g, P < 0.01) compared with non-BPD neonates. From an initial pool (33,252 genes/patient), 4,523 genes exhibited a false discovery rate (FDR) <1%. The area under the receiver operating characteristic curve (AUC) for predicting BPD utilizing gestational age or birth weight was 87.8% and 87.2%, respectively. The machine learning models, using a combination of five genes, revealed AUCs ranging between 85.8% and 96.1%. Pathways integral to T cell development and differentiation were associated with BPD. A derived five-gene whole blood signature can accurately predict BPD in the first week of life.