Machine Learning Algorithm Validation: From Essentials to Advanced Applications and Implications for Regulatory Certification and Deployment

Seeing the random forest through the decision trees. Supporting learning health systems from histopathology with machine learning models: Challenges and opportunities

This paper discusses some overlooked challenges faced when working with machine learning models for histopathology and presents a novel opportunity to support "Learning Health Systems" with them. Initially, the authors elaborate on these challenges after separating them according to their mitigation strategies: those that need innovative approaches, time, or future technological capabilities and those that require a conceptual reappraisal from a critical perspective. Then, a novel opportunity to support "Learning Health Systems" by integrating hidden information extracted by ML models from digitalized histopathology slides with other healthcare big data is presented.

Patient-Centric In Vitro Fertilization Prognostic Counseling Using Machine Learning for the Pragmatist

Although in vitro fertilization (IVF) has become an extremely effective treatment option for infertility, there is significant underutilization of IVF by patients who could benefit from such treatment. In order for patients to choose to consider IVF treatment when appropriate, it is critical for them to be provided with an accurate, understandable IVF prognosis. Machine learning (ML) can meet the challenge of personalized prognostication based on data available prior to treatment. The development, validation, and deployment of ML prognostic models and related patient counseling report delivery require specialized human and platform expertise. This review article takes a pragmatic approach to review relevant reports of IVF prognostic models and draws from extensive experience meeting patients' and providers' needs with the development of data and model pipelines to implement validated ML models at scale, at the point-of-care. Requirements of using ML-based IVF prognostics at point-of-care will be considered alongside clinical ML implementation factors critical for success. Finally, we discuss health, social, and economic objectives that may be achieved by leveraging combined human expertise and ML prognostics to expand fertility care access and advance health and social good.

Translational Connectomics: overview of machine learning in macroscale Connectomics for clinical insights

Connectomics is a neuroscience paradigm focused on noninvasively mapping highly intricate and organized networks of neurons. The advent of neuroimaging has led to extensive mapping of the brain functional and structural connectome on a macroscale level through modalities such as functional and diffusion MRI. In parallel, the healthcare field has witnessed a surge in the application of machine learning and artificial intelligence for diagnostics, especially in imaging. While reviews covering machine learn ing and macroscale connectomics exist for specific disorders, none provide an overview that captures their evolving role, especially through the lens of clinical application and translation. The applications include understanding disorders, classification, identifying neuroimaging biomarkers, assessing severity, predicting outcomes and intervention response, identifying potential targets for brain stimulation, and evaluating the effects of stimulation intervention on the brain and connectome mapping in patients before neurosurgery. The covered studies span neurodegenerative, neurodevelopmental, neuropsychiatric, and neurological disorders. Along with applications, the review provides a brief of common ML methods to set context. Conjointly, limitations in ML studies within connectomics and strategies to mitigate them have been covered.

Cross-attention enables deep learning on limited omics-imaging-clinical data of 130 lung cancer patients

Deep-learning tools that extract prognostic factors derived from multi-omics data have recently contributed to individualized predictions of survival outcomes. However, the limited size of integrated omics-imaging-clinical datasets poses challenges. Here, we propose two biologically interpretable and robust deep-learning architectures for survival prediction of non-small cell lung cancer (NSCLC) patients, learning simultaneously from computed tomography (CT) scan images, gene expression data, and clinical information. The proposed models integrate patient-specific clinical, transcriptomic, and imaging data and incorporate Kyoto Encyclopedia of Genes and Genomes (KEGG) and Reactome pathway information, adding biological knowledge within the learning process to extract prognostic gene biomarkers and molecular pathways. While both models accurately stratify patients in high- and low-risk groups when trained on a dataset of only 130 patients, introducing a cross-attention mechanism in a sparse autoencoder significantly improves the performance, highlighting tumor regions and NSCLC-related genes as potential biomarkers and thus offering a significant methodological advancement when learning from small imaging-omics-clinical samples.

Strategies for Reliable Stress Recognition: A Machine Learning Approach Using Heart Rate Variability Features

Stress recognition, particularly using machine learning (ML) with physiological data such as heart rate variability (HRV), holds promise for mental health interventions. However, limited datasets in affective computing and healthcare research can lead to inaccurate conclusions regarding the ML model performance. This study employed supervised learning algorithms to classify stress and relaxation states using HRV measures. To account for limitations associated with small datasets, robust strategies were implemented based on methodological recommendations for ML with a limited dataset, including data segmentation, feature selection, and model evaluation. Our findings highlight that the random forest model achieved the best performance in distinguishing stress from non-stress states. Notably, it showed higher performance in identifying stress from relaxation (F1-score: 86.3%) compared to neutral states (F1-score: 65.8%). Additionally, the model demonstrated generalizability when tested on independent secondary datasets, showcasing its ability to distinguish between stress and relaxation states. While our performance metrics might be lower than some previous studies, this likely reflects our focus on robust methodologies to enhance the generalizability and interpretability of ML models, which are crucial for real-world applications with limited datasets.

Frameworks for procurement, integration, monitoring, and evaluation of artificial intelligence tools in clinical settings: A systematic review

Research on the applications of artificial intelligence (AI) tools in medicine has increased exponentially over the last few years but its implementation in clinical practice has not seen a commensurate increase with a lack of consensus on implementing and maintaining such tools. This systematic review aims to summarize frameworks focusing on procuring, implementing, monitoring, and evaluating AI tools in clinical practice. A comprehensive literature search, following PRSIMA guidelines was performed on MEDLINE, Wiley Cochrane, Scopus, and EBSCO databases, to identify and include articles recommending practices, frameworks or guidelines for AI procurement, integration, monitoring, and evaluation. From the included articles, data regarding study aim, use of a framework, rationale of the framework, details regarding AI implementation involving procurement, integration, monitoring, and evaluation were extracted. The extracted details were then mapped on to the Donabedian Plan, Do, Study, Act cycle domains. The search yielded 17,537 unique articles, out of which 47 were evaluated for inclusion based on their full texts and 25 articles were included in the review. Common themes extracted included transparency, feasibility of operation within existing workflows, integrating into existing workflows, validation of the tool using predefined performance indicators and improving the algorithm and/or adjusting the tool to improve performance. Among the four domains (Plan, Do, Study, Act) the most common domain was Plan (84%, n = 21), followed by Study (60%, n = 15), Do (52%, n = 13), & Act (24%, n = 6). Among 172 authors, only 1 (0.6%) was from a low-income country (LIC) and 2 (1.2%) were from lower-middle-income countries (LMICs). Healthcare professionals cite the implementation of AI tools within clinical settings as challenging owing to low levels of evidence focusing on integration in the Do and Act domains. The current healthcare AI landscape calls for increased data sharing and knowledge translation to facilitate common goals and reap maximum clinical benefit.

Improved reconstruction for highly accelerated propeller diffusion 1.5 T clinical MRI

PURPOSE

Propeller fast-spin-echo diffusion magnetic resonance imaging (FSE-dMRI) is essential for the diagnosis of Cholesteatoma. However, at clinical 1.5 T MRI, its signal-to-noise ratio (SNR) remains relatively low. To gain sufficient SNR, signal averaging (number of excitations, NEX) is usually used with the cost of prolonged scan time. In this work, we leveraged the benefits of Locally Low Rank (LLR) constrained reconstruction to enhance the SNR. Furthermore, we enhanced both the speed and SNR by employing Convolutional Neural Networks (CNNs) for the accelerated PROPELLER FSE-dMRI on a 1.5 T clinical scanner.

METHODS

Residual U-Net (RU-Net) was found to be efficient for propeller FSE-dMRI data. It was trained to predict 2-NEX images obtained by Locally Low Rank (LLR) constrained reconstruction and used 1-NEX images obtained via simplified reconstruction as the inputs. The brain scans from healthy volunteers and patients with cholesteatoma were performed for model training and testing. The performance of trained networks was evaluated with normalized root-mean-square-error (NRMSE), structural similarity index measure (SSIM), and peak SNR (PSNR).

RESULTS

For 4 × under-sampled with 7 blades data, online reconstruction appears to provide suboptimal images-some small details are missing due to high noise interferences. Offline LLR enables suppression of noises and discovering some small structures. RU-Net demonstrated further improvement compared to LLR by increasing 18.87% of PSNR, 2.11% of SSIM, and reducing 53.84% of NRMSE. Moreover, RU-Net is about 1500 × faster than LLR (0.03 vs. 47.59 s/slice).

CONCLUSION

The LLR remarkably enhances the SNR compared to online reconstruction. Moreover, RU-Net improves propeller FSE-dMRI as reflected in PSNR, SSIM, and NRMSE. It requires only 1-NEX data, which allows a 2 × scan time reduction. In addition, its speed is approximately 1500 times faster than that of LLR-constrained reconstruction.

From big data to big insights: statistical and bioinformatic approaches for exploring the lipidome

The goal of lipidomic studies is to provide a broad characterization of cellular lipids present and changing in a sample of interest. Recent lipidomic research has significantly contributed to revealing the multifaceted roles that lipids play in fundamental cellular processes, including signaling, energy storage, and structural support. Furthermore, these findings have shed light on how lipids dynamically respond to various perturbations. Continued advancement in analytical techniques has also led to improved abilities to detect and identify novel lipid species, resulting in increasingly large datasets. Statistical analysis of these datasets can be challenging not only because of their vast size, but also because of the highly correlated data structure that exists due to many lipids belonging to the same metabolic or regulatory pathways. Interpretation of these lipidomic datasets is also hindered by a lack of current biological knowledge for the individual lipids. These limitations can therefore make lipidomic data analysis a daunting task. To address these difficulties and shed light on opportunities and also weaknesses in current tools, we have assembled this review. Here, we illustrate common statistical approaches for finding patterns in lipidomic datasets, including univariate hypothesis testing, unsupervised clustering, supervised classification modeling, and deep learning approaches. We then describe various bioinformatic tools often used to biologically contextualize results of interest. Overall, this review provides a framework for guiding lipidomic data analysis to promote a greater assessment of lipidomic results, while understanding potential advantages and weaknesses along the way.

The Role of AI in Hospitals and Clinics: Transforming Healthcare in the 21st Century

As healthcare systems around the world face challenges such as escalating costs, limited access, and growing demand for personalized care, artificial intelligence (AI) is emerging as a key force for transformation. This review is motivated by the urgent need to harness AI's potential to mitigate these issues and aims to critically assess AI's integration in different healthcare domains. We explore how AI empowers clinical decision-making, optimizes hospital operation and management, refines medical image analysis, and revolutionizes patient care and monitoring through AI-powered wearables. Through several case studies, we review how AI has transformed specific healthcare domains and discuss the remaining challenges and possible solutions. Additionally, we will discuss methodologies for assessing AI healthcare solutions, ethical challenges of AI deployment, and the importance of data privacy and bias mitigation for responsible technology use. By presenting a critical assessment of AI's transformative potential, this review equips researchers with a deeper understanding of AI's current and future impact on healthcare. It encourages an interdisciplinary dialogue between researchers, clinicians, and technologists to navigate the complexities of AI implementation, fostering the development of AI-driven solutions that prioritize ethical standards, equity, and a patient-centered approach.

AI-organoid integrated systems for biomedical studies and applications

In this review, we explore the growing role of artificial intelligence (AI) in advancing the biomedical applications of human pluripotent stem cell (hPSC)-derived organoids. Stem cell-derived organoids, these miniature organ replicas, have become essential tools for disease modeling, drug discovery, and regenerative medicine. However, analyzing the vast and intricate datasets generated from these organoids can be inefficient and error-prone. AI techniques offer a promising solution to efficiently extract insights and make predictions from diverse data types generated from microscopy images, transcriptomics, metabolomics, and proteomics. This review offers a brief overview of organoid characterization and fundamental concepts in AI while focusing on a comprehensive exploration of AI applications in organoid-based disease modeling and drug evaluation. It provides insights into the future possibilities of AI in enhancing the quality control of organoid fabrication, label-free organoid recognition, and three-dimensional image reconstruction of complex organoid structures. This review presents the challenges and potential solutions in AI-organoid integration, focusing on the establishment of reliable AI model decision-making processes and the standardization of organoid research.

Automated diagnosis of autism with artificial intelligence: State of the art

Autism spectrum disorder (ASD) represents a panel of conditions that begin during the developmental period and result in impairments of personal, social, academic, or occupational functioning. Early diagnosis is directly related to a better prognosis. Unfortunately, the diagnosis of ASD requires a long and exhausting subjective process. We aimed to review the state of the art for automated autism diagnosis and recognition in this research. In February 2022, we searched multiple databases and sources of gray literature for eligible studies. We used an adapted version of the QUADAS-2 tool to assess the risk of bias in the studies. A brief report of the methods and results of each study is presented. Data were synthesized for each modality separately using the Split Component Synthesis (SCS) method. We assessed heterogeneity using the I 2 statistics and evaluated publication bias using trim and fill tests combined with ln DOR. Confidence in cumulative evidence was assessed using the GRADE approach for diagnostic studies. We included 344 studies from 186,020 participants (51,129 are estimated to be unique) for nine different modalities in this review, from which 232 reported sufficient data for meta-analysis. The area under the curve was in the range of 0.71-0.90 for all the modalities. The studies on EEG data provided the best accuracy, with the area under the curve ranging between 0.85 and 0.93. We found that the literature is rife with bias and methodological/reporting flaws. Recommendations are provided for future research to provide better studies and fill in the current knowledge gaps.

External validation of a deep learning algorithm for automated echocardiographic strain measurements

Aims

Echocardiographic strain imaging reflects myocardial deformation and is a sensitive measure of cardiac function and wall-motion abnormalities. Deep learning (DL) algorithms could automate the interpretation of echocardiographic strain imaging.

Methods and results

We developed and trained an automated DL-based algorithm for left ventricular (LV) strain measurements in an internal dataset. Global longitudinal strain (GLS) was validated externally in (i) a real-world Taiwanese cohort of participants with and without heart failure (HF), (ii) a core-lab measured dataset from the multinational prevalence of microvascular dysfunction-HF and preserved ejection fraction (PROMIS-HFpEF) study, and regional strain in (iii) the HMC-QU-MI study of patients with suspected myocardial infarction. Outcomes included measures of agreement [bias, mean absolute difference (MAD), root-mean-squared-error (RMSE), and Pearson's correlation (R)] and area under the curve (AUC) to identify HF and regional wall-motion abnormalities. The DL workflow successfully analysed 3741 (89%) studies in the Taiwanese cohort, 176 (96%) in PROMIS-HFpEF, and 158 (98%) in HMC-QU-MI. Automated GLS showed good agreement with manual measurements (mean ± SD): -18.9 ± 4.5% vs. -18.2 ± 4.4%, respectively, bias 0.68 ± 2.52%, MAD 2.0 ± 1.67, RMSE = 2.61, R = 0.84 in the Taiwanese cohort; and -15.4 ± 4.1% vs. -15.9 ± 3.6%, respectively, bias -0.65 ± 2.71%, MAD 2.19 ± 1.71, RMSE = 2.78, R = 0.76 in PROMIS-HFpEF. In the Taiwanese cohort, automated GLS accurately identified patients with HF (AUC = 0.89 for total HF and AUC = 0.98 for HF with reduced ejection fraction). In HMC-QU-MI, automated regional strain identified regional wall-motion abnormalities with an average AUC = 0.80.

Conclusion

DL algorithms can interpret echocardiographic strain images with similar accuracy as conventional measurements. These results highlight the potential of DL algorithms to democratize the use of cardiac strain measurements and reduce time-spent and costs for echo labs globally.

Bridging artificial intelligence and fucoxanthin for the recovery and quantification from microalgae

Fucoxanthin is a carotenoid that possesses various beneficial medicinal properties for human well-being. However, the current extraction technologies and quantification techniques are still lacking in terms of cost validation, high energy consumption, long extraction time, and low yield production. To date, artificial intelligence (AI) models can assist and improvise the bottleneck of fucoxanthin extraction and quantification process by establishing new technologies and processes which involve big data, digitalization, and automation for efficiency fucoxanthin production. This review highlights the application of AI models such as artificial neural network (ANN) and adaptive neuro fuzzy inference system (ANFIS), capable of learning patterns and relationships from large datasets, capturing non-linearity, and predicting optimal conditions that significantly impact the fucoxanthin extraction yield. On top of that, combining metaheuristic algorithm such as genetic algorithm (GA) can further improve the parameter space and discovery of optimal conditions of ANN and ANFIS models, which results in high R2 accuracy ranging from 98.28% to 99.60% after optimization. Besides, AI models such as support vector machine (SVM), convolutional neural networks (CNNs), and ANN have been leveraged for the quantification of fucoxanthin, either computer vision based on color space of images or regression analysis based on statistical data. The findings are reliable when modeling for the concentration of pigments with high R2 accuracy ranging from 66.0% - 99.2%. This review paper has reviewed the feasibility and potential of AI for the extraction and quantification purposes, which can reduce the cost, accelerate the fucoxanthin yields, and development of fucoxanthin-based products.

Machine Learning for Automated Classification of Abnormal Lung Sounds Obtained from Public Databases: A Systematic Review

Pulmonary auscultation is essential for detecting abnormal lung sounds during physical assessments, but its reliability depends on the operator. Machine learning (ML) models offer an alternative by automatically classifying lung sounds. ML models require substantial data, and public databases aim to address this limitation. This systematic review compares characteristics, diagnostic accuracy, concerns, and data sources of existing models in the literature. Papers published from five major databases between 1990 and 2022 were assessed. Quality assessment was accomplished with a modified QUADAS-2 tool. The review encompassed 62 studies utilizing ML models and public-access databases for lung sound classification. Artificial neural networks (ANN) and support vector machines (SVM) were frequently employed in the ML classifiers. The accuracy ranged from 49.43% to 100% for discriminating abnormal sound types and 69.40% to 99.62% for disease class classification. Seventeen public databases were identified, with the ICBHI 2017 database being the most used (66%). The majority of studies exhibited a high risk of bias and concerns related to patient selection and reference standards. Summarizing, ML models can effectively classify abnormal lung sounds using publicly available data sources. Nevertheless, inconsistent reporting and methodologies pose limitations to advancing the field, and therefore, public databases should adhere to standardized recording and labeling procedures.

Application of machine learning and artificial intelligence in the diagnosis and classification of polycystic ovarian syndrome: a systematic review

Introduction

Polycystic Ovarian Syndrome (PCOS) is the most common endocrinopathy in women of reproductive age and remains widely underdiagnosed leading to significant morbidity. Artificial intelligence (AI) and machine learning (ML) hold promise in improving diagnostics. Thus, we performed a systematic review of literature to identify the utility of AI/ML in the diagnosis or classification of PCOS.

Methods

We applied a search strategy using the following databases MEDLINE, Embase, the Cochrane Central Register of Controlled Trials, the Web of Science, and the IEEE Xplore Digital Library using relevant keywords. Eligible studies were identified, and results were extracted for their synthesis from inception until January 1, 2022.

Results

135 studies were screened and ultimately, 31 studies were included in this study. Data sources used by the AI/ML interventions included clinical data, electronic health records, and genetic and proteomic data. Ten studies (32%) employed standardized criteria (NIH, Rotterdam, or Revised International PCOS classification), while 17 (55%) used clinical information with/without imaging. The most common AI techniques employed were support vector machine (42% studies), K-nearest neighbor (26%), and regression models (23%) were the commonest AI/ML. Receiver operating curves (ROC) were employed to compare AI/ML with clinical diagnosis. Area under the ROC ranged from 73% to 100% (n=7 studies), diagnostic accuracy from 89% to 100% (n=4 studies), sensitivity from 41% to 100% (n=10 studies), specificity from 75% to 100% (n=10 studies), positive predictive value (PPV) from 68% to 95% (n=4 studies), and negative predictive value (NPV) from 94% to 99% (n=2 studies).

Conclusion

Artificial intelligence and machine learning provide a high diagnostic and classification performance in detecting PCOS, thereby providing an avenue for early diagnosis of this disorder. However, AI-based studies should use standardized PCOS diagnostic criteria to enhance the clinical applicability of AI/ML in PCOS and improve adherence to methodological and reporting guidelines for maximum diagnostic utility.

Systematic review registration

https://www.crd.york.ac.uk/prospero/, identifier CRD42022295287.

The Emerging Role of Raman Spectroscopy as an Omics Approach for Metabolic Profiling and Biomarker Detection toward Precision Medicine

Omics technologies have rapidly evolved with the unprecedented potential to shape precision medicine. Novel omics approaches are imperative toallow rapid and accurate data collection and integration with clinical information and enable a new era of healthcare. In this comprehensive review, we highlight the utility of Raman spectroscopy (RS) as an emerging omics technology for clinically relevant applications using clinically significant samples and models. We discuss the use of RS both as a label-free approach for probing the intrinsic metabolites of biological materials, and as a labeled approach where signal from Raman reporters conjugated to nanoparticles (NPs) serve as an indirect measure for tracking protein biomarkers in vivo and for high throughout proteomics. We summarize the use of machine learning algorithms for processing RS data to allow accurate detection and evaluation of treatment response specifically focusing on cancer, cardiac, gastrointestinal, and neurodegenerative diseases. We also highlight the integration of RS with established omics approaches for holistic diagnostic information. Further, we elaborate on metal-free NPs that leverage the biological Raman-silent region overcoming the challenges of traditional metal NPs. We conclude the review with an outlook on future directions that will ultimately allow the adaptation of RS as a clinical approach and revolutionize precision medicine.

Medical imaging and multimodal artificial intelligence models for streamlining and enhancing cancer care: opportunities and challenges

INTRODUCTION

Artificial intelligence (AI) has the potential to transform oncologic care. There have been significant developments in AI applications in medical imaging and increasing interest in multimodal models. These are likely to enable improved oncologic care through more precise diagnosis, increasingly in a more personalized and less invasive manner. In this review, we provide an overview of the current state and challenges that clinicians, administrative personnel and policy makers need to be aware of and mitigate for the technology to reach its full potential.

AREAS COVERED

The article provides a brief targeted overview of AI, a high-level review of the current state and future potential AI applications in diagnostic radiology and to a lesser extent digital pathology, focusing on oncologic applications. This is followed by a discussion of emerging approaches, including multimodal models. The article concludes with a discussion of technical, regulatory challenges and infrastructure needs for AI to realize its full potential.

EXPERT OPINION

There is a large volume of promising research, and steadily increasing commercially available tools using AI. For the most advanced and promising precision diagnostic applications of AI to be used clinically, robust and comprehensive quality monitoring systems and informatics platforms will likely be required.

Small molecule biomarker discovery: Proposed workflow for LC-MS-based clinical research projects

Mass spectrometry focusing on small endogenous molecules has become an integral part of biomarker discovery in the pursuit of an in-depth understanding of the pathophysiology of various diseases, ultimately enabling the application of personalized medicine. While LC-MS methods allow researchers to gather vast amounts of data from hundreds or thousands of samples, the successful execution of a study as part of clinical research also requires knowledge transfer with clinicians, involvement of data scientists, and interactions with various stakeholders. The initial planning phase of a clinical research project involves specifying the scope and design, and engaging relevant experts from different fields. Enrolling subjects and designing trials rely largely on the overall objective of the study and epidemiological considerations, while proper pre-analytical sample handling has immediate implications on the quality of analytical data. Subsequent LC-MS measurements may be conducted in a targeted, semi-targeted, or non-targeted manner, resulting in datasets of varying size and accuracy. Data processing further enhances the quality of data and is a prerequisite for in-silico analysis. Nowadays, the evaluation of such complex datasets relies on a mix of classical statistics and machine learning applications, in combination with other tools, such as pathway analysis and gene set enrichment. Finally, results must be validated before biomarkers can be used as prognostic or diagnostic decision-making tools. Throughout the study, quality control measures should be employed to enhance the reliability of data and increase confidence in the results. The aim of this graphical review is to provide an overview of the steps to be taken when conducting an LC-MS-based clinical research project to search for small molecule biomarkers.

Traffic monitoring system design considering multi-hazard disaster risks

Roadways are critical infrastructure in our society, providing services for people through and between cities. However, they are prone to closures and disruptions, especially after extreme weather events like hurricanes. At the same time, traffic flow data are a fundamental type of information for any transportation system. In this paper, we tackle the problem of traffic sensor placement on roadways to address two tasks at the same time. The first task is traffic data estimation in ordinary situations, which is vital for traffic monitoring and city planning. We design a graph-based method to estimate traffic flow on roads where sensors are not present. The second one is enhanced observability of roadways in case of extreme weather events. We propose a satellite-based multi-domain risk assessment to locate roads at high risk of closures. Vegetation and flood hazards are taken into account. We formalize the problem as a search method over the network to suggest the minimum number and location of traffic sensors to place while maximizing the traffic estimation capabilities and observability of the risky areas of a city.

Introduction to Radiomics and Artificial Intelligence: A Primer for Radiologists

Health informatics and artificial intelligence (AI) are expected to transform the healthcare enterprise and the future practice of radiology. There is an increasing body of literature on radiomics and deep learning/AI applications in medical imaging. There are also a steadily increasing number of FDA cleared AI applications in radiology. It is therefore essential for radiologists to have a basic understanding of these approaches, whether in academia or private practice. In this article, we will provide an overview of the field and familiarize the readers with the fundamental concepts behind these approaches.

A Practical Guide for AI Algorithm Selection for the Radiology Department

There is a steadily increasing number of artificial intelligence (AI) tools available and cleared for use in clinical radiological practice. Radiologists will increasingly be faced with options provided by other radiologist colleagues, clinician colleagues, vendors, or other professionals for obtaining and deploying AI algorithms in clinical practice. It is important that radiologists are familiar with basic and practical aspects that need to be considered when assessing an AI tool for use in their practice, so that resources are properly allocated and there is an appropriate return on investment through enhancements in patient quality of care, safety, and/or process efficiency. In this review, we will discuss a potential approach for AI software assessment and practical points that should be considered when considering the acquisition and deployment of an AI tool in the radiology department.

Trends in digital image processing of isolated microalgae by incorporating classification algorithm

Identification of microalgae species is of importance due to the uprising of harmful algae blooms affecting both the aquatic habitat and human health. Despite this occurence, microalgae have been identified as a green biomass and alternative source due to its promising bioactive compounds accumulation that play a significant role in many industrial applications. Recently, microalgae species identification has been conducted through DNA analysis and various microscopy techniques such as light, scanning electron, transmission electron, and atomic force -microscopy. The aforementioned procedures have encouraged researchers to consider alternate ways due to limitations such as costly validation, requiring skilled taxonomists, prolonged analysis, and low accuracy. This review highlights the potential innovations in digital microscopy with the incorporation of both hardware and software that can produce a reliable recognition, detection, enumeration, and real-time acquisition of microalgae species. Several steps such as image acquisition, processing, feature extraction, and selection are discussed, for the purpose of generating high image quality by removing unwanted artifacts and noise from the background. These steps of identification of microalgae species is performed by reliable image classification through machine learning as well as deep learning algorithms such as artificial neural networks, support vector machines, and convolutional neural networks. Overall, this review provides comprehensive insights into numerous possibilities of microalgae image identification, image pre-processing, and machine learning techniques to address the challenges in developing a robust digital classification tool for the future.

Semi-Self-Supervised Learning for Semantic Segmentation in Images with Dense Patterns

Deep learning has shown potential in domains with large-scale annotated datasets. However, manual annotation is expensive, time-consuming, and tedious. Pixel-level annotations are particularly costly for semantic segmentation in images with dense irregular patterns of object instances, such as in plant images. In this work, we propose a method for developing high-performing deep learning models for semantic segmentation of such images utilizing little manual annotation. As a use case, we focus on wheat head segmentation. We synthesize a computationally annotated dataset-using a few annotated images, a short unannotated video clip of a wheat field, and several video clips with no wheat-to train a customized U-Net model. Considering the distribution shift between the synthesized and real images, we apply three domain adaptation steps to gradually bridge the domain gap. Only using two annotated images, we achieved a Dice score of 0.89 on the internal test set. When further evaluated on a diverse external dataset collected from 18 different domains across five countries, this model achieved a Dice score of 0.73. To expose the model to images from different growth stages and environmental conditions, we incorporated two annotated images from each of the 18 domains to further fine-tune the model. This increased the Dice score to 0.91. The result highlights the utility of the proposed approach in the absence of large-annotated datasets. Although our use case is wheat head segmentation, the proposed approach can be extended to other segmentation tasks with similar characteristics of irregularly repeating patterns of object instances.

Experimental and clinical data analysis for identification of COVID-19 resistant ACE2 mutations

The high magnitude zoonotic event has caused by Severe Acute Respitarory Syndrome CoronaVirus-2 (SARS-CoV-2) is Coronavirus Disease-2019 (COVID-19) epidemics. This disease has high rate of spreading than mortality in humans. The human receptor, Angiotensin-Converting Enzyme 2 (ACE2), is the leading target site for viral Spike-protein (S-protein) that function as binding ligands and are responsible for their entry in humans. The patients infected with COVID-19 with comorbidities, particularly cancer patients, have a severe effect or high mortality rate because of the suppressed immune system. Nevertheless, there might be a chance wherein cancer patients cannot be infected with SARS-CoV-2 because of mutations in the ACE2, which may be resistant to the spillover between species. This study aimed to determine the mutations in the sequence of the human ACE2 protein and its dissociation with SARS-CoV-2 that might be rejecting viral transmission. The in silico approaches were performed to identify the impact of SARS-CoV-2 S-protein with ACE2 mutations, validated experimentally, occurred in the patient, and reported in cell lines. The identified changes significantly affect SARS-CoV-2 S-protein interaction with ACE2, demonstrating the reduction in the binding affinity compared to SARS-CoV. The data presented in this study suggest ACE2 mutants have a higher and lower affinity with SARS-Cov-2 S-protein to the wild-type human ACE2 receptor. This study would likely be used to report SARS-CoV-2 resistant ACE2 mutations and can be used to design active peptide development to inactivate the viral spread of SARS-CoV-2 in humans.

Use of machine learning in pediatric surgical clinical prediction tools: A systematic review

PURPOSE

Clinical prediction tools (CPTs) are decision-making instruments utilizing patient data to predict specific clinical outcomes, risk-stratify patients, or suggest personalized diagnostic or therapeutic options. Recent advancements in artificial intelligence have resulted in a proliferation of CPTs created using machine learning (ML)-yet the clinical applicability of ML-based CPTs and their validation in clinical settings remain unclear. This systematic review aims to compare the validity and clinical efficacy of ML-based to traditional CPTs in pediatric surgery.

METHODS

Nine databases were searched from 2000 until July 9, 2021 to retrieve articles reporting on CPTs and ML for pediatric surgical conditions. PRISMA standards were followed, and screening was performed by two independent reviewers in Rayyan, with a third reviewer resolving conflicts. Risk of bias was assessed using the PROBAST.

RESULTS

Out of 8300 studies, 48 met the inclusion criteria. The most represented surgical specialties were pediatric general (14), neurosurgery (13) and cardiac surgery (12). Prognostic (26) CPTs were the most represented type of surgical pediatric CPTs followed by diagnostic (10), interventional (9), and risk stratifying (2). One study included a CPT for diagnostic, interventional and prognostic purposes. 81% of studies compared their CPT to ML-based CPTs, statistical CPTs, or the unaided clinician, but lacked external validation and/or evidence of clinical implementation.

CONCLUSIONS

While most studies claim significant potential improvements by incorporating ML-based CPTs in pediatric surgical decision-making, both external validation and clinical application remains limited. Further studies must focus on validating existing instruments or developing validated tools, and incorporating them in the clinical workflow.

TYPE OF STUDY

Systematic Review LEVEL OF EVIDENCE: Level III.

A scoping review of natural language processing of radiology reports in breast cancer

Various natural language processing (NLP) algorithms have been applied in the literature to analyze radiology reports pertaining to the diagnosis and subsequent care of cancer patients. Applications of this technology include cohort selection for clinical trials, population of large-scale data registries, and quality improvement in radiology workflows including mammography screening. This scoping review is the first to examine such applications in the specific context of breast cancer. Out of 210 identified articles initially, 44 met our inclusion criteria for this review. Extracted data elements included both clinical and technical details of studies that developed or evaluated NLP algorithms applied to free-text radiology reports of breast cancer. Our review illustrates an emphasis on applications in diagnostic and screening processes over treatment or therapeutic applications and describes growth in deep learning and transfer learning approaches in recent years, although rule-based approaches continue to be useful. Furthermore, we observe increased efforts in code and software sharing but not with data sharing.

IoT Based System for Heart Monitoring and Arrhythmia Detection Using Machine Learning

Internet of Things (IoT) technologies allow building a digital representation of people, objects, or physical phenomena to be available on the Internet. Thus, stakeholders can access this information from remote places or computational systems could analyze this data to find patterns, make decisions, or execute actions. For instance, a doctor could diagnose patients by analyzing the received data from an IoT system even when patients are located in a remote place. This article proposes an IoT system for monitoring electrocardiogram (ECG) signal and processing heart data in order to generate an alert when an arrhythmia is present. This system involves a Polar H10 heart sensor, machine-learning models to classify heart events, and communication technology to share and store patient's information. In the first place, the architecture of the IoT monitoring system and the communication between the components are described by discussing the designing criteria. Second, the experimentation process performs the training and the assessment of three classification algorithms, random forest, convolutional neural network, and k-nearest neighbors. The results show that k-nearest neighbor has the best accuracy percentage classifying the arrhythmias under study (premature ventricular contraction 94%, fusion of ventricular beat 81%, and supraventricular premature beat 82%); also, it is able to discern normal and unclassifiable beats with 93% and 97%, respectively.

Machine learning algorithm as a prognostic tool for venous thromboembolism in allogeneic transplant patients

As a serious complication after allogenic hematopoietic stem cell transplantation (allo-HSCT), venous thromboembolism (VTE) is significantly related to increased nonrelapse mortality. Therefore distinguishing patients at high risk of death who should receive specific therapeutic management is key to improving survival. This study aimed to establish a machine learning-based prognostic model for the identification of post-transplantation VTE patients who have a high risk of death. We retrospectively evaluated 256 consecutive VTE patients who underwent allo-HSCT at our center between 2008 and 2019. These patients were further randomly divided into (1) a derivation (80%) cohort of 205 patients and (2) a test (20%) cohort of 51 patients. The least absolute shrinkage and selection operator (LASSO) approach was used to choose the potential predictors from the primary dataset. Eight machine learning classifiers were used to produce 8 candidate models. A 10-fold cross-validation procedure was used to internally evaluate the models and to select the best-performing model for external assessment using the test cohort. In total, 256 of 7238 patients were diagnosed with VTE after transplantation. Among them, 118 patients (46.1%) had catheter-related venous thrombosis, 107 (41.8%) had isolated deep-vein thrombosis (DVT), 20 (7.8%) had isolated pulmonary embolism (PE), and 11 (4.3%) had concomitant DVT and PE. The 2-year overall survival (OS) rate of patients with VTE was 68.8%. Using LASSO regression, 8 potential features were selected from the 54 candidate variables. The best-performing algorithm based on the 10-fold cross-validation runs was a logistic regression classifier. Therefore a prognostic model named BRIDGE was then established to predict the 2-year OS rate. The areas under the curves of the BRIDGE model were 0.883, 0.871, and 0.858 for the training, validation, and test cohorts, respectively. The Hosmer-Lemeshow goodness-of-fit test showed a high agreement between the predicted and observed outcomes. Decision curve analysis indicated that VTE patients could benefit from the clinical application of the prognostic model. A BRIDGE risk score calculator for predicting the study result is available online (47.94.162.105:8080/bridge/). We established the BRIDGE model to precisely predict the risk for all-cause death in VTE patients after allo-HSCT. Identifying VTE patients who have a high risk of death can help physicians treat these patients in advance, which will improve patient survival.

In silico prediction of siRNA ionizable-lipid nanoparticles In vivo efficacy: Machine learning modeling based on formulation and molecular descriptors

In silico prediction of the in vivo efficacy of siRNA ionizable-lipid nanoparticles is desirable as it can save time and resources dedicated to wet-lab experimentation. This study aims to computationally predict siRNA nanoparticles in vivo efficacy. A data set containing 120 entries was prepared by combining molecular descriptors of the ionizable lipids together with two nanoparticles formulation characteristics. Input descriptor combinations were selected by an evolutionary algorithm. Artificial neural networks, support vector machines and partial least squares regression were used for QSAR modeling. Depending on how the data set is split, two training sets and two external validation sets were prepared. Training and validation sets contained 90 and 30 entries respectively. The results showed the successful predictions of validation set log (siRNA dose) with R_val ²= 0.86-0.89 and 0.75-80 for validation sets one and two, respectively. Artificial neural networks resulted in the best R_val ² for both validation sets. For predictions that have high bias, improvement of R_val ² from 0.47 to 0.96 was achieved by selecting the training set lipids lying within the applicability domain. In conclusion, in vivo performance of siRNA nanoparticles was successfully predicted by combining cheminformatics with machine learning techniques.

Recommendations on compiling test datasets for evaluating artificial intelligence solutions in pathology

Artificial intelligence (AI) solutions that automatically extract information from digital histology images have shown great promise for improving pathological diagnosis. Prior to routine use, it is important to evaluate their predictive performance and obtain regulatory approval. This assessment requires appropriate test datasets. However, compiling such datasets is challenging and specific recommendations are missing. A committee of various stakeholders, including commercial AI developers, pathologists, and researchers, discussed key aspects and conducted extensive literature reviews on test datasets in pathology. Here, we summarize the results and derive general recommendations on compiling test datasets. We address several questions: Which and how many images are needed? How to deal with low-prevalence subsets? How can potential bias be detected? How should datasets be reported? What are the regulatory requirements in different countries? The recommendations are intended to help AI developers demonstrate the utility of their products and to help pathologists and regulatory agencies verify reported performance measures. Further research is needed to formulate criteria for sufficiently representative test datasets so that AI solutions can operate with less user intervention and better support diagnostic workflows in the future.

Prediction of radiation-induced acute skin toxicity in breast cancer patients using data encapsulation screening and dose-gradient-based multi-region radiomics technique: A multicenter study

Purpose

Radiation-induced dermatitis is one of the most common side effects for breast cancer patients treated with radiation therapy (RT). Acute complications can have a considerable impact on tumor control and quality of life for breast cancer patients. In this study, we aimed to develop a novel quantitative high-accuracy machine learning tool for prediction of radiation-induced dermatitis (grade ≥ 2) (RD 2+) before RT by using data encapsulation screening and multi-region dose-gradient-based radiomics techniques, based on the pre-treatment planning computed tomography (CT) images, clinical and dosimetric information of breast cancer patients.

Methods and Materials

214 patients with breast cancer who underwent RT between 2018 and 2021 were retrospectively collected from 3 cancer centers in China. The CT images, as well as the clinical and dosimetric information of patients were retrieved from the medical records. 3 PTV dose related ROIs, including irradiation volume covered by 100%, 105%, and 108% of prescribed dose, combined with 3 skin dose-related ROIs, including irradiation volume covered by 20-Gy, 30-Gy, 40-Gy isodose lines within skin, were contoured for radiomics feature extraction. A total of 4280 radiomics features were extracted from all 6 ROIs. Meanwhile, 29 clinical and dosimetric characteristics were included in the data analysis. A data encapsulation screening algorithm was applied for data cleaning. Multiple-variable logistic regression and 5-fold-cross-validation gradient boosting decision tree (GBDT) were employed for modeling training and validation, which was evaluated by using receiver operating characteristic analysis.

Results

The best predictors for symptomatic RD 2+ were the combination of 20 radiomics features, 8 clinical and dosimetric variables, achieving an area under the curve (AUC) of 0.998 [95% CI: 0.996-1.0] and an AUC of 0.911 [95% CI: 0.838-0.983] in the training and validation dataset, respectively, in the 5-fold-cross-validation GBDT model. Meanwhile, the top 12 most important characteristics as well as their corresponding importance measures for RD 2+ prediction in the GBDT machine learning process were identified and calculated.

Conclusions

A novel multi-region dose-gradient-based GBDT machine learning framework with a random forest based data encapsulation screening method integrated can achieve a high-accuracy prediction of acute RD 2+ in breast cancer patients.

Pivotal challenges in artificial intelligence and machine learning applications for neonatal care

Clinical decision support systems (CDSS) that are developed based on artificial intelligence and machine learning (AI/ML) approaches carry transformative potentials in improving the way neonatal care is practiced. From the use of the data available from electronic health records to physiological sensors and imaging modalities, CDSS can be used to predict clinical outcomes (such as mortality rate, hospital length of state, or surgical outcome) or early warning signs of diseases in neonates. However, only a limited number of clinical decision support systems for neonatal care are currently deployed in healthcare facilities or even implemented during pilot trials (or prospective studies). This is mostly due to the unresolved challenges in developing a real-time supported clinical decision support system, which mainly consists of three phases: model development, model evaluation, and real-time deployment. In this review, we introduce some of the pivotal challenges and factors we must consider during the implementation of real-time supported CDSS.

Importance of sex and gender factors for COVID-19 infection and hospitalisation: a sex-stratified analysis using machine learning in UK Biobank data

OBJECTIVE

To examine sex and gender roles in COVID-19 test positivity and hospitalisation in sex-stratified predictive models using machine learning.

DESIGN

Cross-sectional study.

SETTING

UK Biobank prospective cohort.

PARTICIPANTS

Participants tested between 16 March 2020 and 18 May 2020 were analysed.

MAIN OUTCOME MEASURES

The endpoints of the study were COVID-19 test positivity and hospitalisation. Forty-two individuals' demographics, psychosocial factors and comorbidities were used as likely determinants of outcomes. Gradient boosting machine was used for building prediction models.

RESULTS

Of 4510 individuals tested (51.2% female, mean age=68.5±8.9 years), 29.4% tested positive. Males were more likely to be positive than females (31.6% vs 27.3%, p=0.001). In females, living in more deprived areas, lower income, increased low-density lipoprotein (LDL) to high-density lipoprotein (HDL) ratio, working night shifts and living with a greater number of family members were associated with a higher likelihood of COVID-19 positive test. While in males, greater body mass index and LDL to HDL ratio were the factors associated with a positive test. Older age and adverse cardiometabolic characteristics were the most prominent variables associated with hospitalisation of test-positive patients in both overall and sex-stratified models.

CONCLUSION

High-risk jobs, crowded living arrangements and living in deprived areas were associated with increased COVID-19 infection in females, while high-risk cardiometabolic characteristics were more influential in males. Gender-related factors have a greater impact on females; hence, they should be considered in identifying priority groups for COVID-19 infection vaccination campaigns.

Radiomics and machine learning for the diagnosis of pediatric cervical non-tuberculous mycobacterial lymphadenitis

Non-tuberculous mycobacterial (NTM) infection is an emerging infectious entity that often presents as lymphadenitis in the pediatric age group. Current practice involves invasive testing and excisional biopsy to diagnose NTM lymphadenitis. In this study, we performed a retrospective analysis of 249 lymph nodes selected from 143 CT scans of pediatric patients presenting with lymphadenopathy at the Montreal Children's Hospital between 2005 and 2018. A Random Forest classifier was trained on the ten most discriminative features from a set of 1231 radiomic features. The model classifying nodes as pyogenic, NTM, reactive, or proliferative lymphadenopathy achieved an accuracy of 72%, a precision of 68%, and a recall of 70%. Between NTM and all other causes of lymphadenopathy, the model achieved an area under the curve (AUC) of 89%. Between NTM and pyogenic lymphadenitis, the model achieved an AUC of 90%. Between NTM and the reactive and proliferative lymphadenopathy groups, the model achieved an AUC of 93%. These results indicate that radiomics can achieve a high accuracy for classification of NTM lymphadenitis. Such a non-invasive highly accurate diagnostic approach has the potential to reduce the need for invasive procedures in the pediatric population.

Artificial intelligence model validation before its application in clinical diagnosis assistance

The process of selecting an artificial intelligence (AI) model to assist clinical diagnosis of a particular pathology and its validation tests is relevant since the values of accuracy, sensitivity and specificity may not reflect the behavior of the method in a real environment. Here, we provide helpful considerations to increase the success of using an AI model in clinical practice.

Training Strategies for Radiology Deep Learning Models in Data-limited Scenarios

Data-driven approaches have great potential to shape future practices in radiology. The most straightforward strategy to obtain clinically accurate models is to use large, well-curated and annotated datasets. However, patient privacy constraints, tedious annotation processes, and the limited availability of radiologists pose challenges to building such datasets. This review details model training strategies in scenarios with limited data, insufficiently labeled data, and/or limited expert resources. This review discusses strategies to enlarge the data sample, decrease the time burden of manual supervised labeling, adjust the neural network architecture to improve model performance, apply semisupervised approaches, and leverage efficiencies from pretrained models. Keywords: Computer-aided Detection/Diagnosis, Transfer Learning, Limited Annotated Data, Augmentation, Synthetic Data, Semisupervised Learning, Federated Learning, Few-Shot Learning, Class Imbalance.

Site-Specific Variation in Radiomic Features of Head and Neck Squamous Cell Carcinoma and Its Impact on Machine Learning Models

Simple Summary

Head and neck squamous cell carcinoma (HNSCC) is the most common mucosal malignancy of the head and neck and a leading cause of cancer death. HNSCC arises from different primary anatomical locations that are typically combined during radiomic analyses assuming that the radiomic features, i.e., quantitative image-based features, are similar based on histopathologic characteristics. However, whether these quantitative features are comparable across tumor sites remains unknown. The aim of our retrospective study was to assess if systematic differences exist between radiomic features based on different tumor sites in HNSCC and how they might affect machine learning model performance in endpoint prediction. Using a population of 605 HNSCC patients, we observed significant differences in radiomic features of tumors from different locations and showed that these differences can impact machine learning model performance. This suggests that tumor site should be considered when developing and evaluating radiomics-based models.

Abstract

Current radiomic studies of head and neck squamous cell carcinomas (HNSCC) are typically based on datasets combining tumors from different locations, assuming that the radiomic features are similar based on histopathologic characteristics. However, molecular pathogenesis and treatment in HNSCC substantially vary across different tumor sites. It is not known if a statistical difference exists between radiomic features from different tumor sites and how they affect machine learning model performance in endpoint prediction. To answer these questions, we extracted radiomic features from contrast-enhanced neck computed tomography scans (CTs) of 605 patients with HNSCC originating from the oral cavity, oropharynx, and hypopharynx/larynx. The difference in radiomic features of tumors from these sites was assessed using statistical analyses and Random Forest classifiers on the radiomic features with 10-fold cross-validation to predict tumor sites, nodal metastasis, and HPV status. We found statistically significant differences (p-value ≤ 0.05) between the radiomic features of HNSCC depending on tumor location. We also observed that differences in quantitative features among HNSCC from different locations impact the performance of machine learning models. This suggests that radiomic features may reveal biologic heterogeneity complementary to current gold standard histopathologic evaluation. We recommend considering tumor site in radiomic studies of HNSCC.

Language Impairment in Alzheimer's Disease-Robust and Explainable Evidence for AD-Related Deterioration of Spontaneous Speech Through Multilingual Machine Learning

Alzheimer's disease (AD) is a pervasive neurodegenerative disease that affects millions worldwide and is most prominently associated with broad cognitive decline, including language impairment. Picture description tasks are routinely used to monitor language impairment in AD. Due to the high amount of manual resources needed for an in-depth analysis of thereby-produced spontaneous speech, advanced natural language processing (NLP) combined with machine learning (ML) represents a promising opportunity. In this applied research field though, NLP and ML methodology do not necessarily ensure robust clinically actionable insights into cognitive language impairment in AD and additional precautions must be taken to ensure clinical-validity and generalizability of results. In this study, we add generalizability through multilingual feature statistics to computational approaches for the detection of language impairment in AD. We include 154 participants (78 healthy subjects, 76 patients with AD) from two different languages (106 English speaking and 47 French speaking). Each participant completed a picture description task, in addition to a battery of neuropsychological tests. Each response was recorded and manually transcribed. From this, task-specific, semantic, syntactic and paralinguistic features are extracted using NLP resources. Using inferential statistics, we determined language features, excluding task specific features, that are significant in both languages and therefore represent "generalizable" signs for cognitive language impairment in AD. In a second step, we evaluated all features as well as the generalizable ones for English, French and both languages in a binary discrimination ML scenario (AD vs. healthy) using a variety of classifiers. The generalizable language feature set outperforms the all language feature set in English, French and the multilingual scenarios. Semantic features are the most generalizable while paralinguistic features show no overlap between languages. The multilingual model shows an equal distribution of error in both English and French. By leveraging multilingual statistics combined with a theory-driven approach, we identify AD-related language impairment that generalizes beyond a single corpus or language to model language impairment as a clinically-relevant cognitive symptom. We find a primary impairment in semantics in addition to mild syntactic impairment, possibly confounded by additional impaired cognitive functions.