Clinical decision support systems for improving diagnostic accuracy and achieving precision medicine

A decision support system to follow up and diagnose primary headache patients using semantically enriched data

BACKGROUND

Headache disorders are an important health burden, having a large health-economic impact worldwide. Current treatment & follow-up processes are often archaic, creating opportunities for computer-aided and decision support systems to increase their efficiency. Existing systems are mostly completely data-driven, and the underlying models are a black-box, deteriorating interpretability and transparency, which are key factors in order to be deployed in a clinical setting.

METHODS

In this paper, a decision support system is proposed, composed of three components: (i) a cross-platform mobile application to capture the required data from patients to formulate a diagnosis, (ii) an automated diagnosis support module that generates an interpretable decision tree, based on data semantically annotated with expert knowledge, in order to support physicians in formulating the correct diagnosis and (iii) a web application such that the physician can efficiently interpret captured data and learned insights by means of visualizations.

RESULTS

We show that decision tree induction techniques achieve competitive accuracy rates, compared to other black- and white-box techniques, on a publicly available dataset, referred to as migbase. Migbase contains aggregated information of headache attacks from 849 patients. Each sample is labeled with one of three possible primary headache disorders. We demonstrate that we are able to reduce the classification error, statistically significant (ρ≤0.05), with more than 10% by balancing the dataset using prior expert knowledge. Furthermore, we achieve high accuracy rates by using features extracted using the Weisfeiler-Lehman kernel, which is completely unsupervised. This makes it an ideal approach to solve a potential cold start problem.

CONCLUSION

Decision trees are the perfect candidate for the automated diagnosis support module. They achieve predictive performances competitive to other techniques on the migbase dataset and are, foremost, completely interpretable. Moreover, the incorporation of prior knowledge increases both predictive performance as well as transparency of the resulting predictive model on the studied dataset.

Prospects and challenges for clinical decision support in the era of big data

Recently, there has been burgeoning interest in developing more effective and robust clinical decision support systems (CDSSs) for oncology. This has been primarily driven by the demands for more personalized and precise medical practice in oncology in the era of so-called Big Data (BD); an era that promises to harness the power of large-scale data flow to revolutionize cancer treatment. This interest in BD analytics has created new opportunities as well as new unmet challenges. These include: routine aggregation and standardization of clinical data; patient privacy; transformation of current analytical approaches to handle such noisy and heterogeneous data; and expanded use of advanced statistical learning methods based on confluence of modern statistical methods and machine learning algorithms. In this review, we present the current status of CDSSs in oncology, the prospects and current challenges of BD analytics, and the promising role of integrated modern statistics and machine learning algorithms in predicting complex clinical endpoints, individualizing treatment rules, and optimizing dynamic personalized treatment regimens. We discuss issues pertaining to these topics and present application examples from an aggregate of experiences. We also discuss the role of human factors in improving the utilization and acceptance of such enhanced CDSSs and how to mitigate possible sources of human error to achieve optimal performance and wider acceptance.

FHIRChain: Applying Blockchain to Securely and Scalably Share Clinical Data

Secure and scalable data sharing is essential for collaborative clinical decision making. Conventional clinical data efforts are often siloed, however, which creates barriers to efficient information exchange and impedes effective treatment decision made for patients. This paper provides four contributions to the study of applying blockchain technology to clinical data sharing in the context of technical requirements defined in the “Shared Nationwide Interoperability Roadmap” from the Office of the National Coordinator for Health Information Technology (ONC). First, we analyze the ONC requirements and their implications for blockchain-based systems. Second, we present FHIRChain, which is a blockchain-based architecture designed to meet ONC requirements by encapsulating the HL7 Fast Healthcare Interoperability Resources (FHIR) standard for shared clinical data. Third, we demonstrate a FHIRChain-based decentralized app using digital health identities to authenticate participants in a case study of collaborative decision making for remote cancer care. Fourth, we highlight key lessons learned from our case study.

Evaluating the Efficiency and Safety of Speech Recognition within a Commercial Electronic Health Record System: A Replication Study

OBJECTIVE

To conduct a replication study to validate previously identified significant risks and inefficiencies associated with the use of speech recognition (SR) for documentation within an electronic health record (EHR) system.

METHODS

Thirty-five emergency department clinicians undertook randomly allocated clinical documentation tasks using keyboard and mouse (KBM) or SR using a commercial EHR system. The experiment design, setting, and tasks (E2) replicated an earlier study (E1), while technical integration issues that may have led to poorer SR performance were addressed.

RESULTS

Complex tasks were significantly slower to complete using SR (16.94%) than KBM (KBM: 191.9 s, SR: 224.4 s; p = 0.009; CI, 11.9-48.3), replicating task completion times observed in the earlier experiment. Errors (non-typographical) were significantly higher with SR compared with KBM for both simple (KBM: 3, SR: 84; p < 0.001; CI, 1.5-2.5) and complex tasks (KBM: 23, SR: 53; p = 0.001; CI, 0.5-1.0), again replicating earlier results (E1: 170, E2: 163; p = 0.660; CI, 0.0-0.0). Typographical errors were reduced significantly in the new study (E1: 465, E2: 150; p < 0.001; CI, 2.0-3.0).

DISCUSSION

The results of this study replicate those reported earlier. The use of SR for clinical documentation within an EHR system appears to be consistently associated with decreased time efficiencies and increased errors. Modifications implemented to optimize SR integration in the EHR seem to have resulted in minor improvements that did not fundamentally change overall results.

CONCLUSION

This replication study adds further evidence for the poor performance of SR-assisted clinical documentation within an EHR. Replication studies remain rare in informatics literature, especially where study results are unexpected or have significant implication; such studies are clearly needed to avoid overdependence on the results of a single study.

Improving pathology and laboratory medicine in low-income and middle-income countries: roadmap to solutions

Insufficient awareness of the centrality of pathology and laboratory medicine (PALM) to a functioning health-care system at policy and governmental level, with the resultant inadequate investment, has meant that efforts to enhance PALM in low-income and middle-income countries have been local, fragmented, and mostly unsustainable. Responding to the four major barriers in PALM service delivery that were identified in the first paper of this Series (workforce, infrastructure, education and training, and quality assurance), this second paper identifies potential solutions that can be applied in low-income and middle-income countries (LMICs). Increasing and retaining a quality PALM workforce requires access to mentorship and continuing professional development, task sharing, and the development of short-term visitor programmes. Opportunities to enhance the training of pathologists and allied PALM personnel by increasing and improving education provision must be explored and implemented. PALM infrastructure must be strengthened by addressing supply chain barriers, and ensuring laboratory information systems are in place. New technologies, including telepathology and point-of-care testing, can have a substantial role in PALM service delivery, if used appropriately. We emphasise the crucial importance of maintaining PALM quality and posit that all laboratories in LMICs should participate in quality assurance and accreditation programmes. A potential role for public-private partnerships in filling PALM services gaps should also be investigated. Finally, to deliver these solutions and ensure equitable access to essential services in LMICs, we propose a PALM package focused on these countries, integrated within a nationally tiered laboratory system, as part of an overarching national laboratory strategic plan.

Reasons For Physicians Not Adopting Clinical Decision Support Systems: Critical Analysis

Background

Clinical decision support systems (CDSSs) are an integral component of today’s health information technologies. They assist with interpretation, diagnosis, and treatment. A CDSS can be embedded throughout the patient safety continuum providing reminders, recommendations, and alerts to health care providers. Although CDSSs have been shown to reduce medical errors and improve patient outcomes, they have fallen short of their full potential. User acceptance has been identified as one of the potential reasons for this shortfall.

Objective

The purpose of this paper was to conduct a critical review and task analysis of CDSS research and to develop a new framework for CDSS design in order to achieve user acceptance.

Methods

A critical review of CDSS papers was conducted with a focus on user acceptance. To gain a greater understanding of the problems associated with CDSS acceptance, we conducted a task analysis to identify and describe the goals, user input, system output, knowledge requirements, and constraints from two different perspectives: the machine (ie, the CDSS engine) and the user (ie, the physician).

Results

Favorability of CDSSs was based on user acceptance of clinical guidelines, reminders, alerts, and diagnostic suggestions. We propose two models: (1) the user acceptance and system adaptation design model, which includes optimizing CDSS design based on user needs/expectations, and (2) the input-process-output-engagemodel, which reveals to users the processes that govern CDSS outputs.

Conclusions

This research demonstrates that the incorporation of the proposed models will improve user acceptance to support the beneficial effects of CDSSs adoption. Ultimately, if a user does not accept technology, this not only poses a threat to the use of the technology but can also pose a threat to the health and well-being of patients.

The revolution of personalized psychiatry: will technology make it happen sooner?

Personalized medicine (PM) aims to establish a new approach in clinical decision-making, based upon a patient's individual profile in order to tailor treatment to each patient's characteristics. Although this has become a focus of the discussion also in the psychiatric field, with evidence of its high potential coming from several proof-of-concept studies, nearly no tools have been developed by now that are ready to be applied in clinical practice. In this paper, we discuss recent technological advances that can make a shift toward a clinical application of the PM paradigm. We focus specifically on those technologies that allow both the collection of massive as much as real-time data, i.e., electronic medical records and smart wearable devices, and to achieve relevant predictions using these data, i.e. the application of machine learning techniques.

Artificial Intelligence Approach for Variant Reporting

Purpose

Next-generation sequencing technologies are actively applied in clinical oncology. Bioinformatics pipeline analysis is an integral part of this process; however, humans cannot yet realize the full potential of the highly complex pipeline output. As a result, the decision to include a variant in the final report during routine clinical sign-out remains challenging.

Methods

We used an artificial intelligence approach to capture the collective clinical sign-out experience of six board-certified molecular pathologists to build and validate a decision support tool for variant reporting. We extracted all reviewed and reported variants from our clinical database and tested several machine learning models. We used 10-fold cross-validation for our variant call prediction model, which derives a contiguous prediction score from 0 to 1 (no to yes) for clinical reporting.

Results

For each of the 19,594 initial training variants, our pipeline generates approximately 500 features, which results in a matrix of > 9 million data points. From a comparison of naive Bayes, decision trees, random forests, and logistic regression models, we selected models that allow human interpretability of the prediction score. The logistic regression model demonstrated 1% false negativity and 2% false positivity. The final models' Youden indices were 0.87 and 0.77 for screening and confirmatory cutoffs, respectively. Retraining on a new assay and performance assessment in 16,123 independent variants validated our approach (Youden index, 0.93). We also derived individual pathologist-centric models (virtual consensus conference function), and a visual drill-down functionality allows assessment of how underlying features contributed to a particular score or decision branch for clinical implementation.

Conclusion

Our decision support tool for variant reporting is a practically relevant artificial intelligence approach to harness the next-generation sequencing bioinformatics pipeline output when the complexity of data interpretation exceeds human capabilities.

Evidence-based dentistry for planning restorative treatments: barriers and potential solutions

INTRODUCTION

Evidence-based dentistry (EBD) can help provide the best treatment option for every patient, however, its implementation in restorative dentistry is very limited.

OBJECTIVE

This study aimed at assessing the barriers preventing the implementation of EBD among dental undergraduate and graduate students in Montreal, and explore possible solutions to overcome these barriers.

MATERIALS AND METHODS

A cross-sectional survey was conducted by means of a paper format self-administrated questionnaire distributed among dental students. The survey assessed the barriers and potential solutions for implementation of an evidence-based practice.

RESULTS

Sixty-one students completed the questionnaire. Forty-one percent of respondents found evidence-based literature to be the most reliable source of information for restorative treatment planning, however, only 16% used it. They considered that finding reliable information was difficult and they sometimes encountered conflicting information when consulting different sources. Dental students had positive attitudes towards the need for better access to evidence-based literature to assist learning and decision making in restorative treatment planning and to improve treatment outcomes. Even for dentists trained in EBD, online searching takes too much time, and even though it can provide information of better quality than personal intuition, it might not be enough to identify the best available evidence.

CONCLUSIONS

Even though dental students are aware of the importance of EBD in restorative dentistry they rarely apply the concept, mainly due to time constraints. For this reason, implementation of EBD would probably require faster access to evidence-based knowledge.

Damschroder L, Cooper-DeHoff RM, Denny JC, Ginsburg GS, Guan Y, Horowitz CR, Levy KD, Levy MA, Madden EB, Matheny ME, Pollin TI, Pratt VM, Rosenman M, Voils CI, W. Weitzel K, Wilke RA, Ryanne Wu R, Orlando LA. Challenges and strategies for implementing genomic services in diverse settings: experiences from the Implementing GeNomics In pracTicE (IGNITE) network

BACKGROUND

To realize potential public health benefits from genetic and genomic innovations, understanding how best to implement the innovations into clinical care is important. The objective of this study was to synthesize data on challenges identified by six diverse projects that are part of a National Human Genome Research Institute (NHGRI)-funded network focused on implementing genomics into practice and strategies to overcome these challenges.

METHODS

We used a multiple-case study approach with each project considered as a case and qualitative methods to elicit and describe themes related to implementation challenges and strategies. We describe challenges and strategies in an implementation framework and typology to enable consistent definitions and cross-case comparisons. Strategies were linked to challenges based on expert review and shared themes.

RESULTS

Three challenges were identified by all six projects, and strategies to address these challenges varied across the projects. One common challenge was to increase the relative priority of integrating genomics within the health system electronic health record (EHR). Four projects used data warehousing techniques to accomplish the integration. The second common challenge was to strengthen clinicians' knowledge and beliefs about genomic medicine. To overcome this challenge, all projects developed educational materials and conducted meetings and outreach focused on genomic education for clinicians. The third challenge was engaging patients in the genomic medicine projects. Strategies to overcome this challenge included use of mass media to spread the word, actively involving patients in implementation (e.g., a patient advisory board), and preparing patients to be active participants in their healthcare decisions.

CONCLUSIONS

This is the first collaborative evaluation focusing on the description of genomic medicine innovations implemented in multiple real-world clinical settings. Findings suggest that strategies to facilitate integration of genomic data within existing EHRs and educate stakeholders about the value of genomic services are considered important for effective implementation. Future work could build on these findings to evaluate which strategies are optimal under what conditions. This information will be useful for guiding translation of discoveries to clinical care, which, in turn, can provide data to inform continual improvement of genomic innovations and their applications.

Construction and implications of structural equation modeling network for pediatric cataract: a data mining research of rare diseases

BACKGROUND

The majority of rare diseases are complex diseases caused by a combination of multiple morbigenous factors. However, uncovering the complex etiology and pathogenesis of rare diseases is difficult due to limited clinical resources and conventional statistical methods. This study aims to investigate the interrelationship and the effectiveness of potential factors of pediatric cataract, for the exploration of data mining strategy in the scenarios of rare diseases.

METHODS

We established a pilot rare disease specialized care center to systematically record all information and the entire treatment process of pediatric cataract patients. These clinical records contain the medical history, multiple structural indices, and comprehensive functional metrics. A two-layer structural equation model network was applied, and eight potential factors were filtered and included in the final modeling.

RESULTS

Four risk factors (area, density, location, and abnormal pregnancy experience) and four beneficial factors (axis length, uncorrected visual acuity, intraocular pressure, and age at diagnosis) were identified. Quantifiable results suggested that abnormal pregnancy history may be the principle risk factor among medical history for pediatric cataracts. Moreover, axis length, density, uncorrected visual acuity and age at diagnosis served as the dominant factors and should be emphasized in regular clinical practice.

CONCLUSIONS

This study proposes a generalized evidence-based pattern for rare and complex disease data mining, provides new insights and clinical implications on pediatric cataract, and promotes rare-disease research and prevention to benefit patients.

Artificial intelligence in medicine

Artificial Intelligence (AI) is a general term that implies the use of a computer to model intelligent behavior with minimal human intervention. AI is generally accepted as having started with the invention of robots. The term derives from the Czech word robota, meaning biosynthetic machines used as forced labor. In this field, Leonardo Da Vinci's lasting heritage is today's burgeoning use of robotic-assisted surgery, named after him, for complex urologic and gynecologic procedures. Da Vinci's sketchbooks of robots helped set the stage for this innovation. AI, described as the science and engineering of making intelligent machines, was officially born in 1956. The term is applicable to a broad range of items in medicine such as robotics, medical diagnosis, medical statistics, and human biology-up to and including today's "omics". AI in medicine, which is the focus of this review, has two main branches: virtual and physical. The virtual branch includes informatics approaches from deep learning information management to control of health management systems, including electronic health records, and active guidance of physicians in their treatment decisions. The physical branch is best represented by robots used to assist the elderly patient or the attending surgeon. Also embodied in this branch are targeted nanorobots, a unique new drug delivery system. The societal and ethical complexities of these applications require further reflection, proof of their medical utility, economic value, and development of interdisciplinary strategies for their wider application.

Toward Precision Healthcare: Context and Mathematical Challenges

Precision medicine refers to the idea of delivering the right treatment to the right patient at the right time, usually with a focus on a data-centered approach to this task. In this perspective piece, we use the term "precision healthcare" to describe the development of precision approaches that bridge from the individual to the population, taking advantage of individual-level data, but also taking the social context into account. These problems give rise to a broad spectrum of technical, scientific, policy, ethical and social challenges, and new mathematical techniques will be required to meet them. To ensure that the science underpinning "precision" is robust, interpretable and well-suited to meet the policy, ethical and social questions that such approaches raise, the mathematical methods for data analysis should be transparent, robust, and able to adapt to errors and uncertainties. In particular, precision methodologies should capture the complexity of data, yet produce tractable descriptions at the relevant resolution while preserving intelligibility and traceability, so that they can be used by practitioners to aid decision-making. Through several case studies in this domain of precision healthcare, we argue that this vision requires the development of new mathematical frameworks, both in modeling and in data analysis and interpretation.

Precision Medicine With Point-of-Care Ultrasound: The Future of Personalized Pediatric Emergency Care

The Precision Medicine Initiative spearheaded by the National Institute of Health has pioneered a new model of health care focused on health care delivery that is tailored to an individual. Medical advances have already provided clinicians with the tools to better predict treatment outcomes based on the individual needs of each patient's disease process. Three-dimensional printing allows medical devices and implants to be custom made-to-order. Technological advances in preoperative imaging have augmented the ability for surgeons to plan a specific surgical approach for each patient. In a similar vein, point-of-care ultrasound offers the emergency care provider an opportunity to move beyond protocols and provide precise medical care tailored to the acute needs of each ill or injured emergent patient. In this article, we explore several cutting-edge applications of point-of-care ultrasound that can help providers develop a personalized approach to resuscitation and emergent procedures in pediatrics.

Temporal electronic phenotyping by mining careflows of breast cancer patients

In this work we present a careflow mining approach designed to analyze heterogeneous longitudinal data and to identify phenotypes in a patient cohort. The main idea underlying our approach is to combine methods derived from sequential pattern mining and temporal data mining to derive frequent healthcare histories (careflows) in a population of patients. This approach was applied to an integrated data repository containing clinical and administrative data of more than 4000 breast cancer patients. We used the mined histories to identify sub-cohorts of patients grouped according to healthcare activities pathways, then we characterized these sub-cohorts with clinical data. In this way, we were able to perform temporal electronic phenotyping of electronic health records (EHR) data.

Predictive diagnosis of major depression using NMR-based metabolomics and least-squares support vector machine

BACKGROUND

Major depressive (MD) disorder is a serious psychiatric disorder that can result in suicidal behavior if not treated. The MD diagnosis using a standardized instrument instead of a structured interview will be advantageous for treatment and management of the MD, but so far no such technique exists. We developed an integrated analytical method of NMR-based metabolomics and least squares-support vector machine (LS-SVM) for predictive diagnosis of the MD.

METHODS

The metabolite profiles in clinical plasma samples obtained from 72 depressive patients and 54 healthy subjects were analyzed by NMR spectroscopy. Then, LS-SVM models with different kernels were trained and tested using 80% and 20% of samples, respectively.

RESULTS

We found that the best performance for the MD prediction was achieved by LS-SVM equipped with RBF kernel. Moreover, the predictive performance of the MD using multi-biomarkers was largely improved as compared with that using a single biomarker. In this study, the LS-SVM-RBF using glucose-lipid signaling can achieve the MD prediction with the AUC values of 0.94 (0.89-0.99) in the training set and 0.96 (0.92-1.00) in the test set.

CONCLUSION

The LS-SVM-RBF using glucose-lipid signaling obtained from NMR spectroscopy can be used as an auxiliary diagnostic tool for the MD.

The power of capturing and using information at the point of care

Historically, the process of scientific discovery and the process of clinical delivery have been poorly integrated. As a result of the Decade of Health Information Technology (2004-2014), many of the former barriers to capturing, processing, and retrieving medical information at the point of care have been surmounted. This change has the potential to both transform how new medical discoveries are made and how evidence-based medicine is put into practice. To illustrate the impact of this ongoing change, several examples are provided of institutions which have innovatively used information at the point of care.

Creating clinical decision support systems for respiratory medicine

Clinical decision support systems are vital for advances in improving patient therapeutic care. We share lessons learned from creating two respiratory clinical decisions support systems for ventilating patients in a critical care setting.

Recommendations for antiarrhythmic drugs based on latent semantic analysis with fc-means clustering

In this paper, we propose a novel model for the appropriate recommendation of antiarrhythmic drugs by introducing a fusion of a latent semantic analysis and k-means clustering. Our model not only captures the latent factors between the types of arrhythmia and patients but also has the ability to search a group of patients with similar arrhythmias. The performance studies conducted against the MIT-BIH arrhythmia database show that clinicians accepted 66.67% of the drugs recommended from our model with a balanced f-score of 38.08%. Comparative study with previous approach also confirms the effectiveness of our model.

A systematic review of association studies of common variants associated with idiopathic congenital talipes equinovarus (ICTEV) in humans in the past 30 years

The genetic cause of idiopathic congenital talipes equinovarus (ICTEV) is largely unknown. We performed a systematic review to describe the findings from 21 studies that have examined the genetic variants related to ICTEV, and to evaluate the quality of reporting. We found that ICTEV was positively associated with Hox family genes, collagen family genes, GLI3, N-acetylation genes, T-box family genes, apoptotic pathway genes, and muscle contractile family genes. Negative and controversial results were also discussed, and several genes associated with ICTEV were identified. Due to the limitation of the included studies, rare coding variants should be further investigated, sample size should be enlarged, and candidate genes should be replicated in larger ICTEV populations. Epigenetic study, pathways, chromosome capture, and detailed gene-environment interaction will also allow further elucidation of factors involved in ICTEV pathogenesis and may shed light on diagnosis and timely and accurate interventions.

A New Framework and Prototype Solution for Clinical Decision Support and Research in Genomics and Other Data-intensive Fields of Medicine

Introduction:

In genomics and other fields, it is now possible to capture and store large amounts of data in electronic medical records (EMRs). However, it is not clear if the routine accumulation of massive amounts of (largely uninterpretable) data will yield any health benefits to patients. Nevertheless, the use of large-scale medical data is likely to grow. To meet emerging challenges and facilitate optimal use of genomic data, our institution initiated a comprehensive planning process that addresses the needs of all stakeholders (e.g., patients, families, healthcare providers, researchers, technical staff, administrators). Our experience with this process and a key genomics research project contributed to the proposed framework.

Framework:

We propose a two-pronged Genomic Clinical Decision Support System (CDSS) that encompasses the concept of the “Clinical Mendeliome” as a patient-centric list of genomic variants that are clinically actionable and introduces the concept of the “Archival Value Criterion” as a decision-making formalism that approximates the cost-effectiveness of capturing, storing, and curating genome-scale sequencing data. We describe a prototype Genomic CDSS that we developed as a first step toward implementation of the framework.

Conclusion:

The proposed framework and prototype solution are designed to address the perspectives of stakeholders, stimulate effective clinical use of genomic data, drive genomic research, and meet current and future needs. The framework also can be broadly applied to additional fields, including other ‘-omics’ fields. We advocate for the creation of a Task Force on the Clinical Mendeliome, charged with defining Clinical Mendeliomes and drafting clinical guidelines for their use.