Ethical Machine Learning in Healthcare

The Digital Transformation in Health: How AI Can Improve the Performance of Health Systems

Mobile health has the potential to revolutionize health care delivery and patient engagement. In this work, we discuss how integrating Artificial Intelligence into digital health applications focused on supply chain operation, patient management, and capacity building, among other use cases, can improve the health system and public health performance. We present the Causal Foundry Artificial Intelligence and Reinforcement Learning platform, which allows the delivery of adaptive interventions whose impact can be optimized through experimentation and real-time monitoring. The system can integrate multiple data sources and digital health applications. The flexibility of this platform to connect to various mobile health applications and digital devices, and to send personalized recommendations based on past data and predictions, can significantly improve the impact of digital tools on health system outcomes. The potential for resource-poor settings, where the impact of this approach on health outcomes could be decisive, is discussed. This framework is similarly applicable to improving efficiency in health systems where scarcity is not an issue.

Large language models and synthetic health data: progress and prospects

Objectives

Given substantial obstacles surrounding health data acquisition, high-quality synthetic health data are needed to meet a growing demand for the application of advanced analytics for clinical discovery, prediction, and operational excellence. We highlight how recent advances in large language models (LLMs) present new opportunities for progress, as well as new risks, in synthetic health data generation (SHDG).

Materials and Methods

We synthesized systematic scoping reviews in the SHDG domain, recent LLM methods for SHDG, and papers investigating the capabilities and limits of LLMs.

Results

We summarize the current landscape of generative machine learning models (eg, Generative Adversarial Networks) for SHDG, describe remaining challenges and limitations, and identify how recent LLM approaches can potentially help mitigate them.

Discussion

Six research directions are outlined for further investigation of LLMs for SHDG: evaluation metrics, LLM adoption, data efficiency, generalization, health equity, and regulatory challenges.

Conclusion

LLMs have already demonstrated both high potential and risks in the health domain, and it is important to study their advantages and disadvantages for SHDG.

A Causal Mediation Approach to Account for Interaction of Treatment and Intercurrent Events: Using Hypothetical Strategy

Hypothetical strategy is a common strategy for handling intercurrent events (IEs). No current guideline or study considers treatment-IE interaction to target the estimand in any one IE-handling strategy. Based on the hypothetical strategy, we aimed to (1) assess the performance of three estimators with different considerations for the treatment-IE interaction in a simulation and (2) compare the estimation of these estimators in a real trial. Simulation data were generalized based on realistic clinical trials of Alzheimer's disease. The estimand of interest was the effect of treatment with no IE occurring under the hypothetical strategy. Three estimators, namely, G-estimation with and without interaction and IE-ignored estimation, were compared in scenarios where the treatment-IE interaction effect was set as -50% to 50% of the main effect. Bias was the key performance measure. The real case was derived from a randomized trial of methadone maintenance treatment. Only G-estimation with interaction exhibited unbiased estimations regardless of the existence, direction or magnitude of the treatment-IE interaction in those scenarios. Neglecting the interaction and ignoring the IE would introduce a bias as large as 0.093 and 0.241 (true value, -1.561) if the interaction effect existed. In the real case, compared with G-estimation with interaction, G-estimation without interaction and IE-ignored estimation increased the estimand of interest by 33.55% and 34.36%, respectively. This study highlights the importance of considering treatment-IE interaction in the estimand framework. In practice, it would be better to include the interaction in the estimator by default.

Mitigating Racial Bias in Health Care Algorithms: Improving Fairness in Access to Supportive Housing

Algorithms for guiding health care decisions have come under increasing scrutiny for being unfair to certain racial and ethnic groups. The authors describe their multistep process, using data from 3,465 individuals, to reduce racial and ethnic bias in an algorithm developed to identify state Medicaid beneficiaries experiencing homelessness and chronic health needs who were eligible for coordinated health care and housing supports. Through an iterative process of adjusting inputs, reviewing outputs with diverse stakeholders, and performing quality assurance, the authors developed an algorithm that achieved racial and ethnic parity in the selection of eligible Medicaid beneficiaries.

Perceptions Toward Using Artificial Intelligence and Technology for Asthma Attack Risk Prediction: Qualitative Exploration of Māori Views

BACKGROUND

Asthma is a significant global health issue, impacting over 500,000 individuals in New Zealand and disproportionately affecting Māori communities in New Zealand, who experience worse asthma symptoms and attacks. Digital technologies, including artificial intelligence (AI) and machine learning (ML) models, are increasingly popular for asthma risk prediction. However, these AI models may underrepresent minority ethnic groups and introduce bias, potentially exacerbating disparities.

OBJECTIVE

This study aimed to explore the views and perceptions that Māori have toward using AI and ML technologies for asthma self-management, identify key considerations for developing asthma attack risk prediction models, and ensure Māori are represented in ML models without worsening existing health inequities.

METHODS

Semistructured interviews were conducted with 20 Māori participants with asthma, 3 male and 17 female, aged 18-76 years. All the interviews were conducted one-on-one, except for 1 interview, which was conducted with 2 participants. Altogether, 10 web-based interviews were conducted, while the rest were kanohi ki te kanohi (face-to-face). A thematic analysis was conducted to identify the themes. Further, sentiment analysis was carried out to identify the sentiments using a pretrained Bidirectional Encoder Representations from Transformers model.

RESULTS

We identified four key themes: (1) concerns about AI use, (2) interest in using technology to support asthma, (3) desired characteristics of AI-based systems, and (4) experience with asthma management and opportunities for technology to improve care. AI was relatively unfamiliar to many participants, and some of them expressed concerns about whether AI technology could be trusted, kanohi ki te kanohi interaction, and inadequate knowledge of AI and technology. These concerns are exacerbated by the Māori experience of colonization. Most of the participants were interested in using technology to support their asthma management, and we gained insights into user preferences regarding computer-based health care applications. Participants discussed their experiences, highlighting problems with health care quality and limited access to resources. They also mentioned the factors that trigger their asthma control level.

CONCLUSIONS

The exploration revealed that there is a need for greater information about AI and technology for Māori communities and a need to address trust issues relating to the use of technology. Expectations in relation to computer-based applications for health purposes were expressed. The research outcomes will inform future investigations on AI and technology to enhance the health of people with asthma, in particular those designed for Indigenous populations in New Zealand.

Artificial intelligence governance framework for healthcare

Recent advancements in the field of Artificial Intelligence (AI) provide promising applications of this technology with the aim of solving complex healthcare challenges. These include optimizing operational efficiencies, supporting clinical administrative functions, and improving care outcomes. Numerous AI models are validated in research settings but few make their way into useful applications due to challenges associated with implementation and adoption. In this article, we describe some of these challenges, along with the need for a facilitating entity to safely translate AI systems into practical use. The authors propose a new AI governance framework to enable healthcare organizations with a mechanism to implement and adopt AI systems.

Automatic Recommender System of Development Platforms for Smart Contract-Based Health Care Insurance Fraud Detection Solutions: Taxonomy and Performance Evaluation

BACKGROUND

Health care insurance fraud is on the rise in many ways, such as falsifying information and hiding third-party liability. This can result in significant losses for the medical health insurance industry. Consequently, fraud detection is crucial. Currently, companies employ auditors who manually evaluate records and pinpoint fraud. However, an automated and effective method is needed to detect fraud with the continually increasing number of patients seeking health insurance. Blockchain is an emerging technology and is constantly evolving to meet business needs. With its characteristics of immutability, transparency, traceability, and smart contracts, it demonstrates its potential in the health care domain. In particular, self-executable smart contracts are essential to reduce the costs associated with traditional paradigms, which are mostly manual, while preserving privacy and building trust among health care stakeholders, including the patient and the health insurance networks. However, with the proliferation of blockchain development platform options, selecting the right one for health care insurance can be difficult. This study addressed this void and developed an automated decision map recommender system to select the most effective blockchain platform for insurance fraud detection.

OBJECTIVE

This study aims to develop smart contracts for detecting health care insurance fraud efficiently. Therefore, we provided a taxonomy of fraud scenarios and implemented their detection using a blockchain platform that was suitable for health care insurance fraud detection. To automatically and efficiently select the best platform, we proposed and implemented a decision map-based recommender system. For developing the decision-map, we proposed a taxonomy of 102 blockchain platforms.

METHODS

We developed smart contracts for 12 fraud scenarios that we identified in the literature. We used the top 2 blockchain platforms selected by our proposed decision-making map-based recommender system, which is tailored for health care insurance fraud. The map used our taxonomy of 102 blockchain platforms classified according to their application domains.

RESULTS

The recommender system demonstrated that Hyperledger Fabric was the best blockchain platform for identifying health care insurance fraud. We validated our recommender system by comparing the performance of the top 2 platforms selected by our system. The blockchain platform taxonomy that we created revealed that 59 blockchain platforms are suitable for all application domains, 25 are suitable for financial services, and 18 are suitable for various application domains. We implemented fraud detection based on smart contracts.

CONCLUSIONS

Our decision map recommender system, which was based on our proposed taxonomy of 102 platforms, automatically selected the top 2 platforms, which were Hyperledger Fabric and Neo, for the implementation of health care insurance fraud detection. Our performance evaluation of the 2 platforms indicated that Fabric surpassed Neo in all performance metrics, as depicted by our recommender system. We provided an implementation of fraud detection based on smart contracts.

Exploring bias risks in artificial intelligence and targeted medicines manufacturing

BACKGROUND

Though artificial intelligence holds great value for healthcare, it may also amplify health inequalities through risks of bias. In this paper, we explore bias risks in targeted medicines manufacturing. Targeted medicines manufacturing refers to the act of making medicines targeted to individual patients or to subpopulations of patients within a general group, which can be achieved, for example, by means of cell and gene therapies. These manufacturing processes are increasingly reliant on digitalised systems which can be controlled by artificial intelligence algorithms. Whether and how bias might turn up in the process, however, is uncertain due to the novelty of the development.

METHODS

Examining stakeholder views across bioethics, precision medicine, and artificial intelligence, we document a range of opinions from eleven semi-structured interviews about the possibility of bias in AI-driven targeted therapies manufacturing.

RESULT

Findings show that bias can emerge in upstream (research and development) and downstream (medicine production) processes when manufacturing targeted medicines. However, interviewees emphasized that downstream processes, particularly those not relying on patient or population data, may have lower bias risks. The study also identified a spectrum of bias meanings ranging from negative and ambivalent to positive and productive. Notably, some participants highlighted the potential for certain biases to have productive moral value in correcting health inequalities. This idea of "corrective bias" problematizes the conventional understanding of bias as primarily a negative concept defined by systematic error or unfair outcomes and suggests potential value in capitalizing on biases to help address health inequalities. Our analysis also indicates, however, that the concept of "corrective bias" requires further critical reflection before they can be used to this end.

Machine learning applications in studying mental health among immigrants and racial and ethnic minorities: an exploratory scoping review

BACKGROUND

The use of machine learning (ML) in mental health (MH) research is increasing, especially as new, more complex data types become available to analyze. By examining the published literature, this review aims to explore the current applications of ML in MH research, with a particular focus on its use in studying diverse and vulnerable populations, including immigrants, refugees, migrants, and racial and ethnic minorities.

METHODS

From October 2022 to March 2024, Google Scholar, EMBASE, and PubMed were queried. ML-related, MH-related, and population-of-focus search terms were strung together with Boolean operators. Backward reference searching was also conducted. Included peer-reviewed studies reported using a method or application of ML in an MH context and focused on the populations of interest. We did not have date cutoffs. Publications were excluded if they were narrative or did not exclusively focus on a minority population from the respective country. Data including study context, the focus of mental healthcare, sample, data type, type of ML algorithm used, and algorithm performance were extracted from each.

RESULTS

Ultimately, 13 peer-reviewed publications were included. All the articles were published within the last 6 years, and over half of them studied populations within the US. Most reviewed studies used supervised learning to explain or predict MH outcomes. Some publications used up to 16 models to determine the best predictive power. Almost half of the included publications did not discuss their cross-validation method.

CONCLUSIONS

The included studies provide proof-of-concept for the potential use of ML algorithms to address MH concerns in these special populations, few as they may be. Our review finds that the clinical application of these models for classifying and predicting MH disorders is still under development.

Performance of Machine Learning Suicide Risk Models in an American Indian Population

Importance

Few suicide risk identification tools have been developed specifically for American Indian and Alaska Native populations, even though these populations face the starkest suicide-related inequities.

Objective

To examine the accuracy of existing machine learning models in a majority American Indian population.

Design, Setting, and Participants

This prognostic study used secondary data analysis of electronic health record data collected from January 1, 2017, to December 31, 2021. Existing models from the Mental Health Research Network (MHRN) and Vanderbilt University (VU) were fitted. Models were compared with an augmented screening indicator that included any previous attempt, recent suicidal ideation, or a recent positive suicide risk screen result. The comparison was based on the area under the receiver operating characteristic curve (AUROC). The study was performed in partnership with a tribe and local Indian Health Service (IHS) in the Southwest. All patients were 18 years or older with at least 1 encounter with the IHS unit during the study period. Data were analyzed between October 6, 2022, and July 29, 2024.

Exposures

Suicide attempts or deaths within 90 days.

Main Outcomes and Measures

Model performance was compared based on the ability to distinguish between those with a suicide attempt or death within 90 days of their last IHS visit with those without this outcome.

Results

Of 16 835 patients (mean [SD] age, 40.0 [17.5] years; 8660 [51.4%] female; 14 251 [84.7%] American Indian), 324 patients (1.9%) had at least 1 suicide attempt, and 37 patients (0.2%) died by suicide. The MHRN model had an AUROC value of 0.81 (95% CI, 0.77-0.85) for 90-day suicide attempts, whereas the VU model had an AUROC value of 0.68 (95% CI, 0.64-0.72), and the augmented screening indicator had an AUROC value of 0.66 (95% CI, 0.63-0.70). Calibration was poor for both models but improved after recalibration.

Conclusion and Relevance

This prognostic study found that existing risk identification models for suicide prevention held promise when applied to new contexts and performed better than relying on a combined indictor of a positive suicide risk screen result, history of attempt, and recent suicidal ideation.

Moral Engagement and Disengagement in Health Care AI Development

BACKGROUND

Machine learning (ML) is utilized increasingly in health care, and can pose harms to patients, clinicians, health systems, and the public. In response, regulators have proposed an approach that would shift more responsibility to ML developers for mitigating potential harms. To be effective, this approach requires ML developers to recognize, accept, and act on responsibility for mitigating harms. However, little is known regarding the perspectives of developers themselves regarding their obligations to mitigate harms.

METHODS

We conducted 40 semi-structured interviews with developers of ML predictive analytics applications for health care in the United States.

RESULTS

Participants varied widely in their perspectives on personal responsibility and included examples of both moral engagement and disengagement, albeit in a variety of forms. While most (70%) of participants made a statement indicative of moral engagement, most of these statements reflected an awareness of moral issues, while only a subset of these included additional elements of engagement such as recognizing responsibility, alignment with personal values, addressing conflicts of interests, and opportunities for action. Further, we identified eight distinct categories of moral disengagement reflecting efforts to minimize potential harms or deflect personal responsibility for preventing or mitigating harms.

CONCLUSIONS

These findings suggest possible facilitators and barriers to the development of ethical ML that could act by encouraging moral engagement or discouraging moral disengagement. Regulatory approaches that depend on the ability of ML developers to recognize, accept, and act on responsibility for mitigating harms might have limited success without education and guidance for ML developers about the extent of their responsibilities and how to implement them.

A toolbox for surfacing health equity harms and biases in large language models

Large language models (LLMs) hold promise to serve complex health information needs but also have the potential to introduce harm and exacerbate health disparities. Reliably evaluating equity-related model failures is a critical step toward developing systems that promote health equity. We present resources and methodologies for surfacing biases with potential to precipitate equity-related harms in long-form, LLM-generated answers to medical questions and conduct a large-scale empirical case study with the Med-PaLM 2 LLM. Our contributions include a multifactorial framework for human assessment of LLM-generated answers for biases and EquityMedQA, a collection of seven datasets enriched for adversarial queries. Both our human assessment framework and our dataset design process are grounded in an iterative participatory approach and review of Med-PaLM 2 answers. Through our empirical study, we find that our approach surfaces biases that may be missed by narrower evaluation approaches. Our experience underscores the importance of using diverse assessment methodologies and involving raters of varying backgrounds and expertise. While our approach is not sufficient to holistically assess whether the deployment of an artificial intelligence (AI) system promotes equitable health outcomes, we hope that it can be leveraged and built upon toward a shared goal of LLMs that promote accessible and equitable healthcare.

Culturally Sensitive and Inclusive IBD Care

As the prevalence of inflammatory bowel disease (IBD) increases within historically disadvantaged communities, it is imperative to better understand how intersectionality-defined as the complex, cumulative way in which the effects of multiple forms of discrimination (such as racism, sexism, and classism)-intersects and social determinants of health influence the patient's experiences within the medical system when navigating their disease. Culturally sensitive care is characterized by the ability to deliver patient-centered care that recognizes how the intersectionality of an individual's identities impacts their disease journey. An intentional consideration and sensitivity to this impact play important roles in providing an inclusive and welcoming space for historically disadvantaged individuals living with IBD and will help address health inequity in IBD. Cultural competence implies mastery of care that understands and respects values and beliefs across cultures, while cultural humility involves recognizing the complexity of cultural identity and engaging in an ongoing learning process from individual patient experiences. Heightening our patient care goals from cultural competence to cultural sensitivity allows healthcare professionals and the systems in which they practice to lead with cultural humility as they adopt a more inclusive and humble perspective when caring for patient groups with a diverse array of identities and cultures and to avoid maintaining the status quo of implicit and explicit biases that impede the delivery of quality IBD care. In this article, we review the literature on IBD care in historically disadvantaged communities, address culturally sensitive care, and propose a framework to incorporating cultural humility in IBD practices and research.

Recommendations to promote fairness and inclusion in biomedical AI research and clinical use

OBJECTIVE

Understanding and quantifying biases when designing and implementing actionable approaches to increase fairness and inclusion is critical for artificial intelligence (AI) in biomedical applications.

METHODS

In this Special Communication, we discuss how bias is introduced at different stages of the development and use of AI applications in biomedical sciences and health care. We describe various AI applications and their implications for fairness and inclusion in sections on 1) Bias in Data Source Landscapes, 2) Algorithmic Fairness, 3) Uncertainty in AI Predictions, 4) Explainable AI for Fairness and Equity, and 5) Sociological/Ethnographic Issues in Data and Results Representation.

RESULTS

We provide recommendations to address biases when developing and using AI in clinical applications.

CONCLUSION

These recommendations can be applied to informatics research and practice to foster more equitable and inclusive health care systems and research discoveries.

Leveraging large language models to foster equity in healthcare

OBJECTIVES

Large language models (LLMs) are poised to change care delivery, but their impact on health equity is unclear. While marginalized populations have been historically excluded from early technology developments, LLMs present an opportunity to change our approach to developing, evaluating, and implementing new technologies. In this perspective, we describe the role of LLMs in supporting health equity.

MATERIALS AND METHODS

We apply the National Institute on Minority Health and Health Disparities (NIMHD) research framework to explore the use of LLMs for health equity.

RESULTS

We present opportunities for how LLMs can improve health equity across individual, family and organizational, community, and population health. We describe emerging concerns including biased data, limited technology diffusion, and privacy. Finally, we highlight recommendations focused on prompt engineering, retrieval augmentation, digital inclusion, transparency, and bias mitigation.

CONCLUSION

The potential of LLMs to support health equity depends on making health equity a focus from the start.

Map of epigenetic age acceleration: A worldwide analysis

We present a systematic analysis of epigenetic age acceleration based on by far the largest collection of publicly available DNA methylation data for healthy samples (93 datasets, 23 K samples), focusing on the geographic (25 countries) and ethnic (31 ethnicities) aspects around the world. We employed the most popular epigenetic tools for assessing age acceleration and examined their quality metrics and ability to extrapolate to epigenetic data from different tissue types and age ranges different from the training data of these models. In most cases, the models proved to be inconsistent with each other and showed different signs of age acceleration, with the PhenoAge model tending to systematically underestimate and different versions of the GrimAge model tending to systematically overestimate the age prediction of healthy subjects. Referring to data availability and consistency, most countries and populations are still not represented in GEO, moreover, different datasets use different criteria for determining healthy controls. Because of this, it is difficult to fully isolate the contribution of "geography/environment", "ethnicity" and "healthiness" to epigenetic age acceleration. Among the explored metrics, only the DunedinPACE, which measures aging rate, appears to adequately reflect the standard of living and socioeconomic indicators in countries, although it has a limited application to blood methylation data only. Invariably, by epigenetic age acceleration, males age faster than females in most of the studied countries and populations.

Gaps in the Global Regulatory Frameworks for the Use of Artificial Intelligence (AI) in the Healthcare Services Sector and Key Recommendations

Artificial Intelligence (AI) has shown remarkable potential to revolutionise healthcare by enhancing diagnostics, improving treatment outcomes, and streamlining administrative processes. In the global regulatory landscape, several countries are working on regulating AI in healthcare. There are five key regulatory issues that need to be addressed: (i) data security and protection-measures to cover the "digital health footprints" left unknowingly by patients when they access AI in health services; (ii) data quality-availability of safe and secure data and more open database sources for AI, algorithms, and datasets to ensure equity and prevent demographic bias; (iii) validation of algorithms-mapping of the explainability and causability of the AI system; (iv) accountability-whether this lies with the healthcare professional, healthcare organisation, or the personified AI algorithm; (v) ethics and equitable access-whether fundamental rights of people are met in an ethical manner. Policymakers may need to consider the entire life cycle of AI in healthcare services and the databases that were used for the training of the AI system, along with requirements for their risk assessments to be publicly accessible for effective regulatory oversight. AI services that enhance their functionality over time need to undergo repeated algorithmic impact assessment and must also demonstrate real-time performance. Harmonising regulatory frameworks at the international level would help to resolve cross-border issues of AI in healthcare services.

Race adjustments in clinical algorithms can help correct for racial disparities in data quality

Despite ethical and historical arguments for removing race from clinical algorithms, the consequences of removal remain unclear. Here, we highlight a largely undiscussed consideration in this debate: varying data quality of input features across race groups. For example, family history of cancer is an essential predictor in cancer risk prediction algorithms but is less reliably documented for Black participants and may therefore be less predictive of cancer outcomes. Using data from the Southern Community Cohort Study, we assessed whether race adjustments could allow risk prediction models to capture varying data quality by race, focusing on colorectal cancer risk prediction. We analyzed 77,836 adults with no history of colorectal cancer at baseline. The predictive value of self-reported family history was greater for White participants than for Black participants. We compared two cancer risk prediction algorithms-a race-blind algorithm which included standard colorectal cancer risk factors but not race, and a race-adjusted algorithm which additionally included race. Relative to the race-blind algorithm, the race-adjusted algorithm improved predictive performance, as measured by goodness of fit in a likelihood ratio test (P-value: <0.001) and area under the receiving operating characteristic curve among Black participants (P-value: 0.006). Because the race-blind algorithm underpredicted risk for Black participants, the race-adjusted algorithm increased the fraction of Black participants among the predicted high-risk group, potentially increasing access to screening. More broadly, this study shows that race adjustments may be beneficial when the data quality of key predictors in clinical algorithms differs by race group.

Machine Learning for Patient-Based Real-Time Quality Control (PBRTQC), Analytical and Preanalytical Error Detection in Clinical Laboratory

The rapidly evolving field of machine learning (ML), along with artificial intelligence in a broad sense, is revolutionising many areas of healthcare, including laboratory medicine. The amalgamation of the fields of ML and patient-based real-time quality control (PBRTQC) processes could improve the traditional PBRTQC and error detection algorithms in the laboratory. This narrative review discusses published studies on using ML for the detection of systematic errors, non-systematic errors, and combinations of different types of errors in clinical laboratories. The studies discussed used ML for detecting bias, the requirement for re-calibration, samples contaminated with intravenous fluid or EDTA, delayed sample analysis, wrong-blood-in-tube errors, interference or a combination of different types of errors, by comparing the performance of ML models with human validators or traditional PBRTQC algorithms. Advantages, limitations, the creation of standardised ML models, ethical and regulatory aspects and potential future developments have also been discussed in brief.

Machine learning analysis of serum cholesterol's impact on knee osteoarthritis progression

The controversy surrounding whether serum total cholesterol is a risk factor for the graded progression of knee osteoarthritis (KOA) has prompted this study to develop an authentic prediction model using a machine learning (ML) algorithm. The objective was to investigate whether serum total cholesterol plays a significant role in the progression of KOA. This cross-sectional study utilized data from the public database DRYAD. LASSO regression was employed to identify risk factors associated with the graded progression of KOA. Additionally, six ML algorithms were utilized in conjunction with clinical features and relevant variables to construct a prediction model. The significance and ranking of variables were carefully analyzed. The variables incorporated in the model include JBS3, Diabetes, Hypertension, HDL, TC, BMI, SES, and AGE. Serum total cholesterol emerged as a significant risk factor for the graded progression of KOA in all six ML algorithms used for importance ranking. XGBoost algorithm was based on the combined best performance of the training and validation sets. The ML algorithm enables predictive modeling of risk factors for the progression of the KOA K-L classification and confirms that serum total cholesterol is an important risk factor for the progression of KOA.

Targeting Machine Learning and Artificial Intelligence Algorithms in Health Care to Reduce Bias and Improve Population Health

Policy Points Artificial intelligence (AI) is disruptively innovating health care and surpassing our ability to define its boundaries and roles in health care and regulate its application in legal and ethical ways. Significant progress has been made in governance in the United States and the European Union. It is incumbent on developers, end users, the public, providers, health care systems, and policymakers to collaboratively ensure that we adopt a national AI health strategy that realizes the Quintuple Aim; minimizes race-based medicine; prioritizes transparency, equity, and algorithmic vigilance; and integrates the patient and community voices throughout all aspects of AI development and deployment.

Are clinicians ethically obligated to disclose their use of medical machine learning systems to patients?

It is commonly accepted that clinicians are ethically obligated to disclose their use of medical machine learning systems to patients, and that failure to do so would amount to a moral fault for which clinicians ought to be held accountable. Call this 'the disclosure thesis.' Four main arguments have been, or could be, given to support the disclosure thesis in the ethics literature: the risk-based argument, the rights-based argument, the materiality argument and the autonomy argument. In this article, I argue that each of these four arguments are unconvincing, and therefore, that the disclosure thesis ought to be rejected. I suggest that mandating disclosure may also even risk harming patients by providing stakeholders with a way to avoid accountability for harm that results from improper applications or uses of these systems.

Centralized and Federated Models for the Analysis of Clinical Data

The progress of precision medicine research hinges on the gathering and analysis of extensive and diverse clinical datasets. With the continued expansion of modalities, scales, and sources of clinical datasets, it becomes imperative to devise methods for aggregating information from these varied sources to achieve a comprehensive understanding of diseases. In this review, we describe two important approaches for the analysis of diverse clinical datasets, namely the centralized model and federated model. We compare and contrast the strengths and weaknesses inherent in each model and present recent progress in methodologies and their associated challenges. Finally, we present an outlook on the opportunities that both models hold for the future analysis of clinical data.

Biomedical Data Science, Artificial Intelligence, and Ethics: Navigating Challenges in the Face of Explosive Growth

Advances in biomedical data science and artificial intelligence (AI) are profoundly changing the landscape of healthcare. This article reviews the ethical issues that arise with the development of AI technologies, including threats to privacy, data security, consent, and justice, as they relate to donors of tissue and data. It also considers broader societal obligations, including the importance of assessing the unintended consequences of AI research in biomedicine. In addition, this article highlights the challenge of rapid AI development against the backdrop of disparate regulatory frameworks, calling for a global approach to address concerns around data misuse, unintended surveillance, and the equitable distribution of AI's benefits and burdens. Finally, a number of potential solutions to these ethical quandaries are offered. Namely, the merits of advocating for a collaborative, informed, and flexible regulatory approach that balances innovation with individual rights and public welfare, fostering a trustworthy AI-driven healthcare ecosystem, are discussed.

Machine learning: a new era for cardiovascular pregnancy physiology and cardio-obstetrics research

The maternal cardiovascular system undergoes functional and structural adaptations during pregnancy and postpartum to support increased metabolic demands of offspring and placental growth, labor, and delivery, as well as recovery from childbirth. Thus, pregnancy imposes physiological stress upon the maternal cardiovascular system, and in the absence of an appropriate response it imparts potential risks for cardiovascular complications and adverse outcomes. The proportion of pregnancy-related maternal deaths from cardiovascular events has been steadily increasing, contributing to high rates of maternal mortality. Despite advances in cardiovascular physiology research, there is still no comprehensive understanding of maternal cardiovascular adaptations in healthy pregnancies. Furthermore, current approaches for the prognosis of cardiovascular complications during pregnancy are limited. Machine learning (ML) offers new and effective tools for investigating mechanisms involved in pregnancy-related cardiovascular complications as well as the development of potential therapies. The main goal of this review is to summarize existing research that uses ML to understand mechanisms of cardiovascular physiology during pregnancy and develop prediction models for clinical application in pregnant patients. We also provide an overview of ML platforms that can be used to comprehensively understand cardiovascular adaptations to pregnancy and discuss the interpretability of ML outcomes, the consequences of model bias, and the importance of ethical consideration in ML use.

Privacy-Enhancing Technologies in Biomedical Data Science

The rapidly growing scale and variety of biomedical data repositories raise important privacy concerns. Conventional frameworks for collecting and sharing human subject data offer limited privacy protection, often necessitating the creation of data silos. Privacy-enhancing technologies (PETs) promise to safeguard these data and broaden their usage by providing means to share and analyze sensitive data while protecting privacy. Here, we review prominent PETs and illustrate their role in advancing biomedicine. We describe key use cases of PETs and their latest technical advances and highlight recent applications of PETs in a range of biomedical domains. We conclude by discussing outstanding challenges and social considerations that need to be addressed to facilitate a broader adoption of PETs in biomedical data science.

Exploring Spectrogram-Based Audio Classification for Parkinson's Disease: A Study on Speech Classification and Qualitative Reliability Verification

Patients suffering from Parkinson's disease suffer from voice impairment. In this study, we introduce models to classify normal and Parkinson's patients using their speech. We used an AST (audio spectrogram transformer), a transformer-based speech classification model that has recently outperformed CNN-based models in many fields, and a CNN-based PSLA (pretraining, sampling, labeling, and aggregation), a high-performance model in the existing speech classification field, for the study. This study compares and analyzes the models from both quantitative and qualitative perspectives. First, qualitatively, PSLA outperformed AST by more than 4% in accuracy, and the AUC was also higher, with 94.16% for AST and 97.43% for PSLA. Furthermore, we qualitatively evaluated the ability of the models to capture the acoustic features of Parkinson's through various CAM (class activation map)-based XAI (eXplainable AI) models such as GradCAM and EigenCAM. Based on PSLA, we found that the model focuses well on the muffled frequency band of Parkinson's speech, and the heatmap analysis of false positives and false negatives shows that the speech features are also visually represented when the model actually makes incorrect predictions. The contribution of this paper is that we not only found a suitable model for diagnosing Parkinson's through speech using two different types of models but also validated the predictions of the model in practice.

Artificial intelligence in primary care practice: Qualitative study to understand perspectives on using AI to derive patient social data

OBJECTIVE

To understand the perspectives of primary care clinicians and health system leaders on the use of artificial intelligence (AI) to derive information about patients' social determinants of health.

DESIGN

Qualitative study.

SETTING

Ontario, Canada.

METHODS

Semistructured, 30-minute virtual interviews were conducted with eligible participants across Ontario wherein they were asked about their perceptions of using AI to derive social data for patients. A descriptive content analysis was used to elicit themes from the data.

MAIN FINDINGS

A total of 12 interviews were conducted with 7 family physicians, 3 clinical team members of various health professions, and 2 health system leaders. Five main themes described the current state of social determinants of health information, perceived benefits of and concerns with using AI to derive social data, how participants would want to see and use AI-derived social data, and suggestions for ethical principles that should underpin the development of this AI tool.

CONCLUSION

Most participants were enthusiastic about the possibility of using AI to derive social data for patients in primary care but noted concerns that should be addressed first. These findings can guide the development of AI-based tools for use in primary care settings.

Digital health delivery in respiratory medicine: adjunct, replacement or cause for division?

Digital medicine is already well established in respiratory medicine through remote monitoring digital devices which are used in the day-to-day care of patients with asthma, COPD and sleep disorders. Image recognition software, deployed in thoracic radiology for many applications including lung cancer screening, is another application of digital medicine. Used as clinical decision support, this software will soon become part of day-to-day practice once concerns regarding generalisability have been addressed. Embodied in the electronic health record, digital medicine also plays a substantial role in the day-to-day clinical practice of respiratory medicine. Given the considerable work the electronic health record demands from clinicians, the next tangible impact of digital medicine may be artificial intelligence that aids administration, makes record keeping easier and facilitates better digital communication with patients. Future promises of digital medicine are based on their potential to analyse and characterise the large amounts of digital clinical data that are collected in routine care. Offering the potential to predict outcomes and personalise therapy, there is much to be excited by in this new epoch of innovation. However, these digital tools are by no means a silver bullet. It remains uncertain whether, let alone when, the promises of better models of personalisation and prediction will translate into clinically meaningful and cost-effective products for clinicians.

Causal fairness assessment of treatment allocation with electronic health records

OBJECTIVE

Healthcare continues to grapple with the persistent issue of treatment disparities, sparking concerns regarding the equitable allocation of treatments in clinical practice. While various fairness metrics have emerged to assess fairness in decision-making processes, a growing focus has been on causality-based fairness concepts due to their capacity to mitigate confounding effects and reason about bias. However, the application of causal fairness notions in evaluating the fairness of clinical decision-making with electronic health record (EHR) data remains an understudied domain. This study aims to address the methodological gap in assessing causal fairness of treatment allocation with electronic health records data. In addition, we investigate the impact of social determinants of health on the assessment of causal fairness of treatment allocation.

METHODS

We propose a causal fairness algorithm to assess fairness in clinical decision-making. Our algorithm accounts for the heterogeneity of patient populations and identifies potential unfairness in treatment allocation by conditioning on patients who have the same likelihood to benefit from the treatment. We apply this framework to a patient cohort with coronary artery disease derived from an EHR database to evaluate the fairness of treatment decisions.

RESULTS

Our analysis reveals notable disparities in coronary artery bypass grafting (CABG) allocation among different patient groups. Women were found to be 4.4%-7.7% less likely to receive CABG than men in two out of four treatment response strata. Similarly, Black or African American patients were 5.4%-8.7% less likely to receive CABG than others in three out of four response strata. These results were similar when social determinants of health (insurance and area deprivation index) were dropped from the algorithm. These findings highlight the presence of disparities in treatment allocation among similar patients, suggesting potential unfairness in the clinical decision-making process.

CONCLUSION

This study introduces a novel approach for assessing the fairness of treatment allocation in healthcare. By incorporating responses to treatment into fairness framework, our method explores the potential of quantifying fairness from a causal perspective using EHR data. Our research advances the methodological development of fairness assessment in healthcare and highlight the importance of causality in determining treatment fairness.

Racism is an ethical issue for healthcare artificial intelligence

There is growing attention and evidence that healthcare AI is vulnerable to racial bias. Despite the renewed attention to racism in the United States, racism is often disconnected from the literature on ethical AI. Addressing racism as an ethical issue will facilitate the development of trustworthy and responsible healthcare AI.

Artificial Intelligence in Cardiovascular Care-Part 2: Applications: JACC Review Topic of the Week

Recent artificial intelligence (AI) advancements in cardiovascular care offer potential enhancements in effective diagnosis, treatment, and outcomes. More than 600 U.S. Food and Drug Administration-approved clinical AI algorithms now exist, with 10% focusing on cardiovascular applications, highlighting the growing opportunities for AI to augment care. This review discusses the latest advancements in the field of AI, with a particular focus on the utilization of multimodal inputs and the field of generative AI. Further discussions in this review involve an approach to understanding the larger context in which AI-augmented care may exist, and include a discussion of the need for rigorous evaluation, appropriate infrastructure for deployment, ethics and equity assessments, regulatory oversight, and viable business cases for deployment. Embracing this rapidly evolving technology while setting an appropriately high evaluation benchmark with careful and patient-centered implementation will be crucial for cardiology to leverage AI to enhance patient care and the provider experience.

Enhancing psychiatric rehabilitation outcomes through a multimodal multitask learning model based on BERT and TabNet: An approach for personalized treatment and improved decision-making

Evaluating the rehabilitation status of individuals with serious mental illnesses (SMI) necessitates a comprehensive analysis of multimodal data, including unstructured text records and structured diagnostic data. However, progress in the effective assessment of rehabilitation status remains limited. Our study develops a deep learning model integrating Bidirectional Encoder Representations from Transformers (BERT) and TabNet through a late fusion strategy to enhance rehabilitation prediction, including referral risk, dangerous behaviors, self-awareness, and medication adherence, in patients with SMI. BERT processes unstructured textual data, such as doctor's notes, whereas TabNet manages structured diagnostic information. The model's interpretability function serves to assist healthcare professionals in understanding the model's predictive decisions, improving patient care. Our model exhibited excellent predictive performance for all four tasks, with an accuracy exceeding 0.78 and an area under the curve of 0.70. In addition, a series of tests proved the model's robustness, fairness, and interpretability. This study combines multimodal and multitask learning strategies into a model and applies it to rehabilitation assessment tasks, offering a promising new tool that can be seamlessly integrated with the clinical workflow to support the provision of optimized patient care.

The Risks and Challenges of Artificial Intelligence in Endocrinology

Artificial intelligence (AI) holds the promise of addressing many of the numerous challenges healthcare faces, which include a growing burden of illness, an increase in chronic health conditions and disabilities due to aging and epidemiological changes, higher demand for health services, overworked and burned-out clinicians, greater societal expectations, and rising health expenditures. While technological advancements in processing power, memory, storage, and the abundance of data have empowered computers to handle increasingly complex tasks with remarkable success, AI introduces a variety of meaningful risks and challenges. Among these are issues related to accuracy and reliability, bias and equity, errors and accountability, transparency, misuse, and privacy of data. As AI systems continue to rapidly integrate into healthcare settings, it is crucial to recognize the inherent risks they bring. These risks demand careful consideration to ensure the responsible and safe deployment of AI in healthcare.

Variation in monitoring: Glucose measurement in the ICU as a case study to preempt spurious correlations

OBJECTIVE

Health inequities can be influenced by demographic factors such as race and ethnicity, proficiency in English, and biological sex. Disparities may manifest as differential likelihood of testing which correlates directly with the likelihood of an intervention to address an abnormal finding. Our retrospective observational study evaluated the presence of variation in glucose measurements in the Intensive Care Unit (ICU).

METHODS

Using the MIMIC-IV database (2008-2019), a single-center, academic referral hospital in Boston (USA), we identified adult patients meeting sepsis-3 criteria. Exclusion criteria were diabetic ketoacidosis, ICU length of stay under 1 day, and unknown race or ethnicity. We performed a logistic regression analysis to assess differential likelihoods of glucose measurements on day 1. A negative binomial regression was fitted to assess the frequency of subsequent glucose readings. Analyses were adjusted for relevant clinical confounders, and performed across three disparity proxy axes: race and ethnicity, sex, and English proficiency.

RESULTS

We studied 24,927 patients, of which 19.5% represented racial and ethnic minority groups, 42.4% were female, and 9.8% had limited English proficiency. No significant differences were found for glucose measurement on day 1 in the ICU. This pattern was consistent irrespective of the axis of analysis, i.e. race and ethnicity, sex, or English proficiency. Conversely, subsequent measurement frequency revealed potential disparities. Specifically, males (incidence rate ratio (IRR) 1.06, 95% confidence interval (CI) 1.01 - 1.21), patients who identify themselves as Hispanic (IRR 1.11, 95% CI 1.01 - 1.21), or Black (IRR 1.06, 95% CI 1.01 - 1.12), and patients being English proficient (IRR 1.08, 95% CI 1.01 - 1.15) had higher chances of subsequent glucose readings.

CONCLUSION

We found disparities in ICU glucose measurements among patients with sepsis, albeit the magnitude was small. Variation in disease monitoring is a source of data bias that may lead to spurious correlations when modeling health data.

The Importance of Conducting Practice-oriented Research with Underserved Populations

There has been a growing emphasis on dissemination of empirically supported treatments. Dissemination, however, should not be restricted to treatment. It can and, in the spirit of the scientific-practitioner model, should also involve research. Because it focuses on the investigation of clinical routine as it takes place in local settings and because it can involve the collaboration of several stakeholders, practice-oriented research (POR) can be viewed as an optimal research method to be disseminated. POR has the potential of addressing particularly relevant gaps of knowledge and action when implemented in regions of the world that have limited resources for or experiences with empirical research, and/or in clinical settings that are serving clinical populations who are not typically receiving optimal mental care services - specifically, individuals in rural and inner cities that have limited economic and social resources. The establishment and maintenance of POR in such regions and/or settings, however, come with specific obstacles and challenges. Integrating the experiences acquired from research conducted in various continents (Africa, Europe, Latin America, and North America), the goal of this paper is to describe some of these challenges, strategies that have been implemented to address them, as well as new possible directions to facilitate the creation and growth of POR. It also describes how these challenges and ways to deal with them can provide helpful lessons for already existing POR infrastructures.

Current perspectives on the use of artificial intelligence in critical patient safety

Intensive Care Units (ICUs) have undergone enhancements in patient safety, and artificial intelligence (AI) emerges as a disruptive technology offering novel opportunities. While the published evidence is limited and presents methodological issues, certain areas show promise, such as decision support systems, detection of adverse events, and prescription error identification. The application of AI in safety may pursue predictive or diagnostic objectives. Implementing AI-based systems necessitates procedures to ensure secure assistance, addressing challenges including trust in such systems, biases, data quality, scalability, and ethical and confidentiality considerations. The development and application of AI demand thorough testing, encompassing retrospective data assessments, real-time validation with prospective cohorts, and efficacy demonstration in clinical trials. Algorithmic transparency and explainability are essential, with active involvement of clinical professionals being crucial in the implementation process.

Evaluating Algorithmic Bias in 30-Day Hospital Readmission Models: Retrospective Analysis

BACKGROUND

The adoption of predictive algorithms in health care comes with the potential for algorithmic bias, which could exacerbate existing disparities. Fairness metrics have been proposed to measure algorithmic bias, but their application to real-world tasks is limited.

OBJECTIVE

This study aims to evaluate the algorithmic bias associated with the application of common 30-day hospital readmission models and assess the usefulness and interpretability of selected fairness metrics.

METHODS

We used 10.6 million adult inpatient discharges from Maryland and Florida from 2016 to 2019 in this retrospective study. Models predicting 30-day hospital readmissions were evaluated: LACE Index, modified HOSPITAL score, and modified Centers for Medicare & Medicaid Services (CMS) readmission measure, which were applied as-is (using existing coefficients) and retrained (recalibrated with 50% of the data). Predictive performances and bias measures were evaluated for all, between Black and White populations, and between low- and other-income groups. Bias measures included the parity of false negative rate (FNR), false positive rate (FPR), 0-1 loss, and generalized entropy index. Racial bias represented by FNR and FPR differences was stratified to explore shifts in algorithmic bias in different populations.

RESULTS

The retrained CMS model demonstrated the best predictive performance (area under the curve: 0.74 in Maryland and 0.68-0.70 in Florida), and the modified HOSPITAL score demonstrated the best calibration (Brier score: 0.16-0.19 in Maryland and 0.19-0.21 in Florida). Calibration was better in White (compared to Black) populations and other-income (compared to low-income) groups, and the area under the curve was higher or similar in the Black (compared to White) populations. The retrained CMS and modified HOSPITAL score had the lowest racial and income bias in Maryland. In Florida, both of these models overall had the lowest income bias and the modified HOSPITAL score showed the lowest racial bias. In both states, the White and higher-income populations showed a higher FNR, while the Black and low-income populations resulted in a higher FPR and a higher 0-1 loss. When stratified by hospital and population composition, these models demonstrated heterogeneous algorithmic bias in different contexts and populations.

CONCLUSIONS

Caution must be taken when interpreting fairness measures' face value. A higher FNR or FPR could potentially reflect missed opportunities or wasted resources, but these measures could also reflect health care use patterns and gaps in care. Simply relying on the statistical notions of bias could obscure or underplay the causes of health disparity. The imperfect health data, analytic frameworks, and the underlying health systems must be carefully considered. Fairness measures can serve as a useful routine assessment to detect disparate model performances but are insufficient to inform mechanisms or policy changes. However, such an assessment is an important first step toward data-driven improvement to address existing health disparities.

Inherent Bias in Electronic Health Records: A Scoping Review of Sources of Bias

Objectives

1.1Biases inherent in electronic health records (EHRs), and therefore in medical artificial intelligence (AI) models may significantly exacerbate health inequities and challenge the adoption of ethical and responsible AI in healthcare. Biases arise from multiple sources, some of which are not as documented in the literature. Biases are encoded in how the data has been collected and labeled, by implicit and unconscious biases of clinicians, or by the tools used for data processing. These biases and their encoding in healthcare records undermine the reliability of such data and bias clinical judgments and medical outcomes. Moreover, when healthcare records are used to build data-driven solutions, the biases are further exacerbated, resulting in systems that perpetuate biases and induce healthcare disparities. This literature scoping review aims to categorize the main sources of biases inherent in EHRs.

Methods

1.2We queried PubMed and Web of Science on January 19th, 2023, for peer-reviewed sources in English, published between 2016 and 2023, using the PRISMA approach to stepwise scoping of the literature. To select the papers that empirically analyze bias in EHR, from the initial yield of 430 papers, 27 duplicates were removed, and 403 studies were screened for eligibility. 196 articles were removed after the title and abstract screening, and 96 articles were excluded after the full-text review resulting in a final selection of 116 articles.

Results

1.3Systematic categorizations of diverse sources of bias are scarce in the literature, while the effects of separate studies are often convoluted and methodologically contestable. Our categorization of published empirical evidence identified the six main sources of bias: a) bias arising from past clinical trials; b) data-related biases arising from missing, incomplete information or poor labeling of data; human-related bias induced by c) implicit clinician bias, d) referral and admission bias; e) diagnosis or risk disparities bias and finally, (f) biases in machinery and algorithms.

Conclusions

1.4Machine learning and data-driven solutions can potentially transform healthcare delivery, but not without limitations. The core inputs in the systems (data and human factors) currently contain several sources of bias that are poorly documented and analyzed for remedies. The current evidence heavily focuses on data-related biases, while other sources are less often analyzed or anecdotal. However, these different sources of biases add to one another exponentially. Therefore, to understand the issues holistically we need to explore these diverse sources of bias. While racial biases in EHR have been often documented, other sources of biases have been less frequently investigated and documented (e.g. gender-related biases, sexual orientation discrimination, socially induced biases, and implicit, often unconscious, human-related cognitive biases). Moreover, some existing studies lack causal evidence, illustrating the different prevalences of disease across groups, which does not per se prove the causality. Our review shows that data-, human- and machine biases are prevalent in healthcare and they significantly impact healthcare outcomes and judgments and exacerbate disparities and differential treatment. Understanding how diverse biases affect AI systems and recommendations is critical. We suggest that researchers and medical personnel should develop safeguards and adopt data-driven solutions with a "bias-in-mind" approach. More empirical evidence is needed to tease out the effects of different sources of bias on health outcomes.

Application of machine learning for lung cancer survival prognostication-A systematic review and meta-analysis

Introduction

Machine learning (ML) techniques have gained increasing attention in the field of healthcare, including predicting outcomes in patients with lung cancer. ML has the potential to enhance prognostication in lung cancer patients and improve clinical decision-making. In this systematic review and meta-analysis, we aimed to evaluate the performance of ML models compared to logistic regression (LR) models in predicting overall survival in patients with lung cancer.

Methods

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement. A comprehensive search was conducted in Medline, Embase, and Cochrane databases using a predefined search query. Two independent reviewers screened abstracts and conflicts were resolved by a third reviewer. Inclusion and exclusion criteria were applied to select eligible studies. Risk of bias assessment was performed using predefined criteria. Data extraction was conducted using the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies (CHARMS) checklist. Meta-analytic analysis was performed to compare the discriminative ability of ML and LR models.

Results

The literature search resulted in 3,635 studies, and 12 studies with a total of 211,068 patients were included in the analysis. Six studies reported confidence intervals and were included in the meta-analysis. The performance of ML models varied across studies, with C-statistics ranging from 0.60 to 0.85. The pooled analysis showed that ML models had higher discriminative ability compared to LR models, with a weighted average C-statistic of 0.78 for ML models compared to 0.70 for LR models.

Conclusion

Machine learning models show promise in predicting overall survival in patients with lung cancer, with superior discriminative ability compared to logistic regression models. However, further validation and standardization of ML models are needed before their widespread implementation in clinical practice. Future research should focus on addressing the limitations of the current literature, such as potential bias and heterogeneity among studies, to improve the accuracy and generalizability of ML models for predicting outcomes in patients with lung cancer. Further research and development of ML models in this field may lead to improved patient outcomes and personalized treatment strategies.

Machine learning for healthcare that matters: Reorienting from technical novelty to equitable impact

Despite significant technical advances in machine learning (ML) over the past several years, the tangible impact of this technology in healthcare has been limited. This is due not only to the particular complexities of healthcare, but also due to structural issues in the machine learning for healthcare (MLHC) community which broadly reward technical novelty over tangible, equitable impact. We structure our work as a healthcare-focused echo of the 2012 paper "Machine Learning that Matters", which highlighted such structural issues in the ML community at large, and offered a series of clearly defined "Impact Challenges" to which the field should orient itself. Drawing on the expertise of a diverse and international group of authors, we engage in a narrative review and examine issues in the research background environment, training processes, evaluation metrics, and deployment protocols which act to limit the real-world applicability of MLHC. Broadly, we seek to distinguish between machine learning ON healthcare data and machine learning FOR healthcare-the former of which sees healthcare as merely a source of interesting technical challenges, and the latter of which regards ML as a tool in service of meeting tangible clinical needs. We offer specific recommendations for a series of stakeholders in the field, from ML researchers and clinicians, to the institutions in which they work, and the governments which regulate their data access.

Drop the shortcuts: image augmentation improves fairness and decreases AI detection of race and other demographics from medical images

BACKGROUND

It has been shown that AI models can learn race on medical images, leading to algorithmic bias. Our aim in this study was to enhance the fairness of medical image models by eliminating bias related to race, age, and sex. We hypothesise models may be learning demographics via shortcut learning and combat this using image augmentation.

METHODS

This study included 44,953 patients who identified as Asian, Black, or White (mean age, 60.68 years ±18.21; 23,499 women) for a total of 194,359 chest X-rays (CXRs) from MIMIC-CXR database. The included CheXpert images comprised 45,095 patients (mean age 63.10 years ±18.14; 20,437 women) for a total of 134,300 CXRs were used for external validation. We also collected 1195 3D brain magnetic resonance imaging (MRI) data from the ADNI database, which included 273 participants with an average age of 76.97 years ±14.22, and 142 females. DL models were trained on either non-augmented or augmented images and assessed using disparity metrics. The features learned by the models were analysed using task transfer experiments and model visualisation techniques.

FINDINGS

In the detection of radiological findings, training a model using augmented CXR images was shown to reduce disparities in error rate among racial groups (-5.45%), age groups (-13.94%), and sex (-22.22%). For AD detection, the model trained with augmented MRI images was shown 53.11% and 31.01% reduction of disparities in error rate among age and sex groups, respectively. Image augmentation led to a reduction in the model's ability to identify demographic attributes and resulted in the model trained for clinical purposes incorporating fewer demographic features.

INTERPRETATION

The model trained using the augmented images was less likely to be influenced by demographic information in detecting image labels. These results demonstrate that the proposed augmentation scheme could enhance the fairness of interpretations by DL models when dealing with data from patients with different demographic backgrounds.

FUNDING

National Science and Technology Council (Taiwan), National Institutes of Health.

Systemic lupus in the era of machine learning medicine

Artificial intelligence and machine learning applications are emerging as transformative technologies in medicine. With greater access to a diverse range of big datasets, researchers are turning to these powerful techniques for data analysis. Machine learning can reveal patterns and interactions between variables in large and complex datasets more accurately and efficiently than traditional statistical methods. Machine learning approaches open new possibilities for studying SLE, a multifactorial, highly heterogeneous and complex disease. Here, we discuss how machine learning methods are rapidly being integrated into the field of SLE research. Recent reports have focused on building prediction models and/or identifying novel biomarkers using both supervised and unsupervised techniques for understanding disease pathogenesis, early diagnosis and prognosis of disease. In this review, we will provide an overview of machine learning techniques to discuss current gaps, challenges and opportunities for SLE studies. External validation of most prediction models is still needed before clinical adoption. Utilisation of deep learning models, access to alternative sources of health data and increased awareness of the ethics, governance and regulations surrounding the use of artificial intelligence in medicine will help propel this exciting field forward.

Migraine headache (MH) classification using machine learning methods with data augmentation

Migraine headache, a prevalent and intricate neurovascular disease, presents significant challenges in its clinical identification. Existing techniques that use subjective pain intensity measures are insufficiently accurate to make a reliable diagnosis. Even though headaches are a common condition with poor diagnostic specificity, they have a significant negative influence on the brain, body, and general human function. In this era of deeply intertwined health and technology, machine learning (ML) has emerged as a crucial force in transforming every aspect of healthcare, utilizing advanced facilities ML has shown groundbreaking achievements related to developing classification and automatic predictors. With this, deep learning models, in particular, have proven effective in solving complex problems spanning computer vision and data analytics. Consequently, the integration of ML in healthcare has become vital, especially in developing countries where limited medical resources and lack of awareness prevail, the urgent need to forecast and categorize migraines using artificial intelligence (AI) becomes even more crucial. By training these models on a publicly available dataset, with and without data augmentation. This study focuses on leveraging state-of-the-art ML algorithms, including support vector machine (SVM), K-nearest neighbors (KNN), random forest (RF), decision tree (DST), and deep neural networks (DNN), to predict and classify various types of migraines. The proposed models with data augmentations were trained to classify seven various types of migraine. The proposed models with data augmentations were trained to classify seven various types of migraine. The revealed results show that DNN, SVM, KNN, DST, and RF achieved an accuracy of 99.66%, 94.60%, 97.10%, 88.20%, and 98.50% respectively with data augmentation highlighting the transformative potential of AI in enhancing migraine diagnosis.