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

Ensuring electronic medical record simulation through better training, modeling, and evaluation

OBJECTIVE

Electronic medical records (EMRs) can support medical research and discovery, but privacy risks limit the sharing of such data on a wide scale. Various approaches have been developed to mitigate risk, including record simulation via generative adversarial networks (GANs). While showing promise in certain application domains, GANs lack a principled approach for EMR data that induces subpar simulation. In this article, we improve EMR simulation through a novel pipeline that (1) enhances the learning model, (2) incorporates evaluation criteria for data utility that informs learning, and (3) refines the training process.

MATERIALS AND METHODS

We propose a new electronic health record generator using a GAN with a Wasserstein divergence and layer normalization techniques. We designed 2 utility measures to characterize similarity in the structural properties of real and simulated EMRs in the original and latent space, respectively. We applied a filtering strategy to enhance GAN training for low-prevalence clinical concepts. We evaluated the new and existing GANs with utility and privacy measures (membership and disclosure attacks) using billing codes from over 1 million EMRs at Vanderbilt University Medical Center.

RESULTS

The proposed model outperformed the state-of-the-art approaches with significant improvement in retaining the nature of real records, including prediction performance and structural properties, without sacrificing privacy. Additionally, the filtering strategy achieved higher utility when the EMR training dataset was small.

CONCLUSIONS

These findings illustrate that EMR simulation through GANs can be substantially improved through more appropriate training, modeling, and evaluation criteria.

Barriers and opportunities for the treatment of mild-to-moderate depression with a watchful waiting approach

OBJECTIVE

The aim of this study is to explore barriers and opportunities in non-pharmacological treatment of depression in primary care (PC) from the perspective of family physicians (FPs).

METHODS

Qualitative analysis was used to explore a sample of 36 FPs treating patients with depressive symptoms. Criteria to maximize variability were followed. Participants were identified by key informants. Six group interviews were developed following a semi-structured thematic script. All interviews were transcribed, analyzed and triangulated. Information was saturated. Principals of reflexivity and circularity were implemented.

RESULTS

The results obtained followed 3 main theoretical axes: the FP, the patient, the healthcare system, and the interaction between them. Barriers included poor alignment with clinical practice guidelines, inadequate FP training, patients' preferences and structural challenges in PC. Among opportunities were good FP clinical interview skills, the beneficial bond of trust between patients and FPs and improved communication with mental healthcare services.

CONCLUSION

Based on FPs' perceptions, non-pharmacological treatment of depression in PC is particularly limited by lack of structured training; patients' preferences and treatment expectations; structural challenges in PC; and insufficient support from specialized mental health professionals.

PRACTICE IMPLICATIONS

Resources for education, structural support in PC and modified back up from mental healthcare services are needed.

Lessons learnt from CBR practice at Hua Don Primary Health Care, Thailand

Purpose

The purpose of this study is to determine the capacity map of professional learning community (PLC) practicing community-based research (CBR) in Ubon Ratchathani Rajabhat University, Thailand, and the implementation of the lessons learnt from the process and essential skills at Hua Don Primary Health Care (PHC).

Design/methodology/approach

Participatory action research (PAR) design was conducted in two phases, one on campus and the other in the PHC. For gathering and validating the data, the snowball sampling technique, focus group, in-depth interviews and the triangulation method were used.

Findings

The PLC capacity map from the first phase provided the essential skills of CBR and the second phase revealed lessons learnt from the implementation in the Hua Don PHC. The shortcut in researching a new target area by a collaboration of the community leader and village health volunteers was prominent. The results could be interpreted in creating collaboration in health care with a new community.

Originality/value

The capacity map is a practical guideline for a beginner or CBR novice researcher, and the lessons learnt help the implementation in the health field, particularly in PHC, succeed smoothly.

An Integrated Platform for Skin Cancer Heterogenous and Multilayered Data Management

Electronic health record (EHR) systems improve health care services by allowing the combination of health data with clinical decision support features and clinical image analyses. This study presents a modular and distributed platform that is able to integrate and accommodate heterogeneous, multidimensional (omics, histological images and clinical) data for the multi-angled portrayal and management of skin cancer patients. The proposed design offers a layered analytical framework as an expansion of current EHR systems, which can integrate high-volume molecular -omics data, imaging data, as well as relevant clinical observations. We present a case study in the field of dermatology, where we attempt to combine the multilayered information for the early detection and characterization of melanoma. The specific architecture aspires to lower the barrier for the introduction of personalized therapeutic approaches, towards precision medicine. The paper describes the technical issues of implementation, along with an initial evaluation of the system and discussion.

Progress of Artificial Intelligence in Gynecological Malignant Tumors

Artificial intelligence (AI) is a sort of new technical science which can simulate, extend and expand human intelligence by developing theories, methods and application systems. In the last five years, the application of AI in medical research has become a hot topic in modern science and technology. Gynecological malignant tumors involves a wide range of knowledge, and AI can play an important part in these aspects, such as medical image recognition, auxiliary diagnosis, drug research and development, treatment scheme formulation and other fields. The purpose of this paper is to describe the progress of AI in gynecological malignant tumors and discuss some problems in its application. It is believed that AI improves the efficiency of diagnosis, reduces the burden of clinicians, and improves the effect of treatment and prognosis. AI will play an irreplaceable role in the field of gynecological malignant oncology and will promote the development of medicine and further promote the transformation from traditional medicine to precision medicine and preventive medicine. However, there are also some problems in the application of AI in gynecologic malignant tumors. For example, AI, inseparable from human participation, still needs to be more “humanized”, and needs to further protect patients’ privacy and health, improve legal and insurance protection, and further improve according to local ethnic conditions and national conditions. However, it is believed that with the continuous development of AI, especially ensemble classifier, and deep learning will have a profound influence on the future of medical technology, which is a powerful driving force for future medical innovation and reform.

Hematology Reference Intervals for Healthy Adults of the City of Sulaymaniyah, Iraq

Background

Hematology laboratory analyses are essential in the diagnosis of and making decisions on clinical cases. Hematology results are only meaningful when reference made to a normal range of values for a particular population. These values are affected by race, diet, and lifestyle the society.

Objective

To establish the hematological reference values for adult residents of the city of Sulaymaniyah, Iraq.

Methods

Blood samples collected from 1133 healthy males and female volunteers were analyzed for complete blood count, serum iron, and vitamin B12 concentrations. After applying the exclusion criteria, the hematology results from 762 individuals comprising 313 males and 449 males were included in the study.

Results

The mean red blood cell count, hemoglobin concentration, hematocrit, and serum iron concentration were higher in males than females while the neutrophil and platelet counts and plateletcrit were higher in females than males.

Conclusion

This study for the first time recorded hematological reference intervals for residents of the city of Sulaymaniyah.

Integration of the Extreme Gradient Boosting model with electronic health records to enable the early diagnosis of multiple sclerosis

BACKGROUND

Delayed multiple sclerosis (MS) diagnoses are not uncommon, an early diagnostic tool is urgently warranted. We aimed to develop an effective tool through electronic health records and machine learning techniques to early recognize MS patients from hospital visitors in China.

METHODS

Two case sets were collected from January 2016 to December 2018. The training set had 239 MS and 1142 controls, and the test set had 23 MS and 92 controls. The utility of Extreme Gradient Boosting (XGBoost), Random Forest (RF), Naive Bayes, K-nearest-neighbor (KNN) and Support Vector Machine (SVM) in early diagnosis of MS was evaluated by the area under curve of receiver operating characteristic, precision, recall, specificity, accuracy and F1 score.

RESULTS

The XGBoost performed the best and was used to generate the results. Thirty-four variables which were highly relevant to MS diagnosis were set for the XGBoost model, and their relative importance with MS were ranked. The training set recall was 0.632, with a precision of 0.576, and the test set recall was 0.609, with a precision of 0.609. Our study found that 61%, 51%, and 49% of the patients could be diagnosed with MS, 1, 2, and 3 years earlier than their real diagnostic time point, respectively.

CONCLUSIONS

A diagnostic tool for early MS recognition based on the XGBoost model and electronic health records were developed to help reduce diagnostic delays in MS.

Proposed Implementation of Blockchain in British Columbia's Health Care Data Management

Background

There are several challenges such as information silos and lack of interoperability with the current electronic medical record (EMR) infrastructure in the Canadian health care system. These challenges can be alleviated by implementing a blockchain-based health care data management solution.

Objective

This study aims to provide a detailed overview of the current health data management infrastructure in British Columbia for identifying some of the gaps and inefficiencies in the Canadian health care data management system. We explored whether blockchain is a viable option for bridging the existing gaps in EMR solutions in British Columbia’s health care system.

Methods

We constructed the British Columbia health care data infrastructure and health information flow based on publicly available information and in partnership with an industry expert familiar with the health systems information technology network of British Columbia’s Provincial Health Services Authorities. Information flow gaps, inconsistencies, and inefficiencies were the target of our analyses.

Results

We found that hospitals and clinics have several choices for managing electronic records of health care information, such as different EMR software or cloud-based data management, and that the system development, implementation, and operations for EMRs are carried out by the private sector. As of 2013, EMR adoption in British Columbia was at 80% across all hospitals and the process of entering medical information into EMR systems in British Columbia could have a lag of up to 1 month. During this lag period, disease progression updates are continually written on physical paper charts and not immediately updated in the system, creating a continuous lag period and increasing the probability of errors and disjointed notes. The current major stumbling block for health care data management is interoperability resulting from the use of a wide range of unique information systems by different health care facilities.

Conclusions

Our analysis of British Columbia’s health care data management revealed several challenges, including information silos, the potential for medical errors, the general unwillingness of parties within the health care system to trust and share data, and the potential for security breaches and operational issues in the current EMR infrastructure. A blockchain-based solution has the highest potential in solving most of the challenges in managing health care data in British Columbia and other Canadian provinces.

Artificial intelligence in in vitro fertilization: a computer decision support system for day-to-day management of ovarian stimulation during in vitro fertilization

OBJECTIVE

To describe a computer algorithm designed for in vitro fertilization (IVF) management and to assess the algorithm's accuracy in the day-to-day decision making during ovarian stimulation for IVF when compared to evidence-based decisions by the clinical team.

DESIGN

Descriptive and comparative study of new technology.

SETTING

Private fertility practice.

INTERVENTION(S)

None.

PATIENT(S)

Data were derived from monitoring during ovarian stimulation from IVF cycles. The database consisted of 2,603 cycles (1,853 autologous and 750 donor cycles) incorporating 7,376 visits for training. An additional 556 unique cycles were used for challenge and to calculate accuracy. There were 59,706 data points. Input variables included estradiol concentrations in picograms per milliliter; ultrasound measurements of follicle diameters in two dimensions in millimeters; cycle day during stimulation and dose of recombinant follicle-stimulating hormone during ovarian stimulation for IVF.

MAIN OUTCOME MEASURE(S)

Accuracy of the algorithm to predict four critical clinical decisions during ovarian stimulation for IVF: [1] stop stimulation or continue stimulation. If the decision was to stop, then the next automated decision was to [2] trigger or cancel. If the decision was to return, then the next key decisions were [3] number of days to follow-up and [4] whether any dosage adjustment was needed.

RESULT(S)

Algorithm accuracies for these four decisions are as follows: continue or stop treatment: 0.92; trigger and schedule oocyte retrieval or cancel cycle: 0.96; dose of medication adjustment: 0.82; and number of days to follow-up: 0.87. These accuracies are for first iteration of the algorithm.

CONCLUSION(S)

We describe a first iteration of a predictive analytic algorithm that is highly accurate and in agreement with evidence-based decisions by expert teams during ovarian stimulation during IVF. These tools offer a potential platform to optimize clinical decision making during IVF.

Nutrition Information in Oncology - Extending the Electronic Patient-Record Data Set

Electronic health records (EHRs) present extensive patient information and may be used as a tool to improve health care. However, the oncology context presents a complex content that increases the difficulties of EHR application. This study aimed at developing openEHR-archetypes representing clinical concepts in cancer nutrition-care, as well as to develop an openEHR-template including the aforementioned archetypes. The study involved the following stages: 1) a thorough literature review, followed by an expert’s (nutrition guideline authors) survey, aiming to identify the main statements of published clinical guidelines on nutrition in cancer patients that were not included on the Clinical Knowledge Manager (CKM) repository; 2) modelling of the archetypes using the Ocean Archetype Software and submission to the CKM repository; 3) creating an example template with Template Designer; and 4) automatic conversion of the openEHR-template into a readily usable EHR using VCIntegrator. The clinical concepts (among 17 clinical concepts not yet available in the CKM repository) chosen for further development were: body composition, diet plan, dietary nutrients, dietary supplements, dietary intake assessment, and Malnutrition Screening Tool (MST). So far, four archetypes were accepted for review in the CKM repository and a template was created and converted into an EHR. This study designed new openEHR-archetypes for nutrition management in cancer patients. These archetypes can be included in EHR. Future studies are needed to assess their applicability in other areas and their practical impact on data quality, system interoperability and, ultimately, on clinical practice and research.

Radiomics-Based Prediction of Overall Survival in Lung Cancer Using Different Volumes-Of-Interest

Lung cancer accounts for the largest amount of deaths worldwide with respect to the other oncological pathologies. To guarantee the most effective cure to patients for such aggressive tumours, radiomics is increasing as a novel and promising research field that aims at extracting knowledge from data in terms of quantitative measures that are computed from diagnostic images, with prognostic and predictive ends. This knowledge could be used to optimize current treatments and to maximize their efficacy. To this end, we hereby study the use of such quantitative biomarkers computed from CT images of patients affected by Non-Small Cell Lung Cancer to predict Overall Survival. The main contributions of this work are two: first, we consider different volumes of interest for the same patient to find out whether the volume surrounding the visible lesions can provide useful information; second, we introduce 3D Local Binary Patterns, which are texture measures scarcely explored in radiomics. As further validation, we show that the proposed signature outperforms not only the features automatically computed by a deep learning-based approach, but also another signature at the state-of-the-art using other handcrafted features.

A Systematic Review of Clinical Decision Support Systems for Clinical Oncology Practice

BACKGROUND

Electronic health records are central to cancer care delivery. Electronic clinical decision support (CDS) systems can potentially improve cancer care quality and safety. However, little is known regarding the use of CDS systems in clinical oncology and their impact on patient outcomes.

METHODS

A systematic review of peer-reviewed studies was performed to evaluate clinically relevant outcomes related to the use of CDS tools for the diagnosis, treatment, and supportive care of patients with cancer. Peer-reviewed studies published from 1995 through 2016 were included if they assessed clinical outcomes, patient-reported outcomes (PROs), costs, or care delivery process measures.

RESULTS

Electronic database searches yielded 2,439 potentially eligible papers, with 24 studies included after final review. Most studies used an uncontrolled, pre-post intervention design. A total of 23 studies reported improvement in key study outcomes with use of oncology CDS systems, and 12 studies assessing the systems for computerized chemotherapy order entry demonstrated reductions in prescribing error rates, medication-related safety events, and workflow interruptions. The remaining studies examined oncology clinical pathways, guideline adherence, systems for collection and communication of PROs, and prescriber alerts.

CONCLUSIONS

There is a paucity of data evaluating clinically relevant outcomes of CDS system implementation in oncology care. Currently available data suggest that these systems can have a positive impact on the quality of cancer care delivery. However, there is a critical need to rigorously evaluate CDS systems in oncology to better understand how they can be implemented to improve patient outcomes.

Artificial intelligence, radiomics and other horizons in body composition assessment

This paper offers a brief overview of common non-invasive techniques for body composition assessment methods, and of the way images extracted by these methods can be processed with artificial intelligence (AI) and radiomic analysis. These new techniques are becoming more and more appealing in the field of health care, thanks to their ability to treat and process a huge amount of data, suggest new correlations between extracted imaging biomarkers and traits of several diseases as well as lead to the possibility to realise an increasingly personalized medicine. The idea is to suggest the use of AI applications and radiomic analysis to search for features that may be extracted from medical images [computed tomography (CT) and magnetic resonance imaging (MRI)], and that may turn out to be good predictors of metabolic disorder diseases and cancer. This could lead to patient-specific treatments and management of several diseases linked with excessive body fat.

Evaluating the Impact of the Grading and Assessment of Predictive Tools Framework on Clinicians and Health Care Professionals' Decisions in Selecting Clinical Predictive Tools: Randomized Controlled Trial

BACKGROUND

While selecting predictive tools for implementation in clinical practice or for recommendation in clinical guidelines, clinicians and health care professionals are challenged with an overwhelming number of tools. Many of these tools have never been implemented or evaluated for comparative effectiveness. To overcome this challenge, the authors developed and validated an evidence-based framework for grading and assessment of predictive tools (the GRASP framework). This framework was based on the critical appraisal of the published evidence on such tools.

OBJECTIVE

The aim of the study was to examine the impact of using the GRASP framework on clinicians' and health care professionals' decisions in selecting clinical predictive tools.

METHODS

A controlled experiment was conducted through a web-based survey. Participants were randomized to either review the derivation publications, such as studies describing the development of the predictive tools, on common traumatic brain injury predictive tools (control group) or to review an evidence-based summary, where each tool had been graded and assessed using the GRASP framework (intervention group). Participants in both groups were asked to select the best tool based on the greatest validation or implementation. A wide group of international clinicians and health care professionals were invited to participate in the survey. Task completion time, rate of correct decisions, rate of objective versus subjective decisions, and level of decisional conflict were measured.

RESULTS

We received a total of 194 valid responses. In comparison with not using GRASP, using the framework significantly increased correct decisions by 64%, from 53.7% to 88.1% (88.1/53.7=1.64; t₁₉₃=8.53; P<.001); increased objective decision making by 32%, from 62% (3.11/5) to 82% (4.10/5; t₁₈₉=9.24; P<.001); decreased subjective decision making based on guessing by 20%, from 49% (2.48/5) to 39% (1.98/5; t₁₈₈=-5.47; P<.001); and decreased prior knowledge or experience by 8%, from 71% (3.55/5) to 65% (3.27/5; t₁₈₇=-2.99; P=.003). Using GRASP significantly decreased decisional conflict and increased the confidence and satisfaction of participants with their decisions by 11%, from 71% (3.55/5) to 79% (3.96/5; t₁₈₈=4.27; P<.001), and by 13%, from 70% (3.54/5) to 79% (3.99/5; t₁₈₈=4.89; P<.001), respectively. Using GRASP decreased the task completion time, on the 90th percentile, by 52%, from 12.4 to 6.4 min (t₁₉₃=-0.87; P=.38). The average System Usability Scale of the GRASP framework was very good: 72.5% and 88% (108/122) of the participants found the GRASP useful.

CONCLUSIONS

Using GRASP has positively supported and significantly improved evidence-based decision making. It has increased the accuracy and efficiency of selecting predictive tools. GRASP is not meant to be prescriptive; it represents a high-level approach and an effective, evidence-based, and comprehensive yet simple and feasible method to evaluate, compare, and select clinical predictive tools.

Using artificial intelligence to improve medical services in China

Artificial intelligence (AI) is one hotspot of research in the field of modern medical technology. Medical AI has been applied to various areas and has two main branches including virtual and physical. Recently, Chinese State Council issued a guideline on developing AI and indicated that the widespread application of AI will improve the level of precision in medical services and achieve the intelligent medical care. Medical resources, especially the high-quality resources, are deficient across the entire health service system in China. AI technologies, such that virtual AI and telemedical technology, are expected to overcome the current limitations of the distribution of medical resources and relieve the pressure associated with obtaining high-quality health care.

Differences in cohort study data affect external validation of artificial intelligence models for predictive diagnostics of dementia - lessons for translation into clinical practice

Artificial intelligence (AI) approaches pose a great opportunity for individualized, pre-symptomatic disease diagnosis which plays a key role in the context of personalized, predictive, and finally preventive medicine (PPPM). However, to translate PPPM into clinical practice, it is of utmost importance that AI-based models are carefully validated. The validation process comprises several steps, one of which is testing the model on patient-level data from an independent clinical cohort study. However, recruitment criteria can bias statistical analysis of cohort study data and impede model application beyond the training data. To evaluate whether and how data from independent clinical cohort studies differ from each other, this study systematically compares the datasets collected from two major dementia cohorts, namely, the Alzheimer's Disease Neuroimaging Initiative (ADNI) and AddNeuroMed. The presented comparison was conducted on individual feature level and revealed significant differences among both cohorts. Such systematic deviations can potentially hamper the generalizability of results which were based on a single cohort dataset. Despite identified differences, validation of a previously published, ADNI trained model for prediction of personalized dementia risk scores on 244 AddNeuroMed subjects was successful: External validation resulted in a high prediction performance of above 80% area under receiver operator characteristic curve up to 6 years before dementia diagnosis. Propensity score matching identified a subset of patients from AddNeuroMed, which showed significantly smaller demographic differences to ADNI. For these patients, an even higher prediction performance was achieved, which demonstrates the influence systematic differences between cohorts can have on validation results. In conclusion, this study exposes challenges in external validation of AI models on cohort study data and is one of the rare cases in the neurology field in which such external validation was performed. The presented model represents a proof of concept that reliable models for personalized predictive diagnostics are feasible, which, in turn, could lead to adequate disease prevention and hereby enable the PPPM paradigm in the dementia field.

Role of Health Information Technology in Addressing Health Disparities: Patient, Clinician, and System Perspectives

Over the last decade, health information technology (IT) has dramatically transformed medical practice in the United States. On May 11-12, 2017, the National Institute on Minority Health and Health Disparities, in partnership with the National Science Foundation and the National Health IT Collaborative for the Underserved, convened a scientific workshop, "Addressing Health Disparities with Health Information Technology," with the goal of ensuring that future research guides potential health IT initiatives to address the needs of health disparities populations. The workshop examined patient, clinician, and system perspectives on the potential role of health IT in addressing health disparities. Attendees were asked to identify and discuss various health IT challenges that confront underserved communities and propose innovative strategies to address them, and to involve these communities in this process. Community engagement, cultural competency, and patient-centered care were highlighted as key to improving health equity, as well as to promoting scalable, sustainable, and effective health IT interventions. Participants noted the need for more research on how health IT can be used to evaluate and address the social determinants of health. Expanding public-private partnerships was emphasized, as was the importance of clinicians and IT developers partnering and using novel methods to learn how to improve health care decision-making. Finally, to advance health IT and promote health equity, it will be necessary to record and capture health disparity data using standardized terminology, and to continuously identify system-level deficiencies and biases.

Weighing the odds: Assessing underdiagnosis of adult obesity via electronic medical record problem list omissions

Background

Obesity is a continuing national epidemic, and the condition can have a physical, psychological, as well as social impact on one’s well-being. Consequently, it is critical to diagnose and document obesity accurately in the patient’s electronic medical record (EMR), so that the information can be used and shared to improve clinical decision making and health communication and, in turn, the patient’s prognosis. It is therefore worthwhile identifying the various factors that play a role in documenting obesity diagnosis and the methods to improve current documentation practices.

Method

We used a retrospective cross-sectional design to analyze outpatient EMRs of patients at an academic outpatient clinic. Obese patients were identified using the measured body mass index (BMI; ≥30 kg/m²) entry in the EMR, recorded at each visit, and checked for documentation of obesity in the EMR problem list. Patients were categorized into two groups (diagnosed or undiagnosed) based on a documented diagnosis (or omission) of obesity in the EMR problem list and compared.

Results

A total of 10,208 unique patient records of obese patients were included for analysis, of which 4119 (40%) did not have any documentation of obesity in their problem list. Chi-square analysis between the diagnosed and undiagnosed groups revealed significant associations between documentation of obesity in the EMR and patient characteristics.

Conclusion

EMR designers and developers must consider employing automated decision support techniques to populate and update problem lists based on the existing recorded BMI in the EMR in order to prevent omissions occurring from manual entry.

Primer on an ethics of AI-based decision support systems in the clinic

Making good decisions in extremely complex and difficult processes and situations has always been both a key task as well as a challenge in the clinic and has led to a large amount of clinical, legal and ethical routines, protocols and reflections in order to guarantee fair, participatory and up-to-date pathways for clinical decision-making. Nevertheless, the complexity of processes and physical phenomena, time as well as economic constraints and not least further endeavours as well as achievements in medicine and healthcare continuously raise the need to evaluate and to improve clinical decision-making. This article scrutinises if and how clinical decision-making processes are challenged by the rise of so-called artificial intelligence-driven decision support systems (AI-DSS). In a first step, this article analyses how the rise of AI-DSS will affect and transform the modes of interaction between different agents in the clinic. In a second step, we point out how these changing modes of interaction also imply shifts in the conditions of trustworthiness, epistemic challenges regarding transparency, the underlying normative concepts of agency and its embedding into concrete contexts of deployment and, finally, the consequences for (possible) ascriptions of responsibility. Third, we draw first conclusions for further steps regarding a 'meaningful human control' of clinical AI-DSS.

Precision health: A pragmatic approach to understanding and addressing key factors in autoimmune diseases

The past decade has witnessed a significant paradigm shift in the clinical approach to autoimmune diseases, lead primarily by initiatives in precision medicine, precision health and precision public health initiatives. An understanding and pragmatic implementation of these approaches require an understanding of the drivers, gaps and limitations of precision medicine. Gaining the trust of the public and patients is paramount but understanding that technologies such as artificial intelligences and machine learning still require context that can only be provided by human input or what is called augmented machine learning. The role of genomics, the microbiome and proteomics, such as autoantibody testing, requires continuing refinement through research and pragmatic approaches to their use in applied precision medicine.

"Balancing Expectations with Actual Realities": Conversations with Clinicians and Scientists in the First Year of a High-Risk Childhood Cancer Precision Medicine Trial

Precision medicine is changing cancer care and placing new demands on oncology professionals. Precision medicine trials for high-risk childhood cancer exemplify these complexities. We assessed clinicians' (n = 39) and scientists' (n = 15) experiences in the first year of the PRecISion Medicine for Children with Cancer (PRISM) trial for children and adolescents with high-risk cancers, through an in-depth semi-structured interview. We thematically analysed participants' responses regarding their professional challenges, and measured oncologists' knowledge of genetics and confidence with somatic and germline molecular test results. Both groups described positive early experiences with PRISM but were cognisant of managing parents' expectations. Key challenges for clinicians included understanding and communicating genomic results, balancing biopsy risks, and drug access. Most oncologists rated 'good' knowledge of genetics, but a minority were 'very confident' in interpreting (25%), explaining (34.4%) and making treatment recommendations (18.8%) based on somatic genetic test results. Challenges for scientists included greater emotional impact of their work and balancing translational outputs with academic productivity. Continued tracking of these challenges across the course of the trial, while assessing the perspectives of a wider range of stakeholders, is critical to drive the ongoing development of a workforce equipped to manage the demands of paediatric precision medicine.

Systematic review of using medical informatics in lung transplantation studies

BACKGROUND

Lung transplantation is one of the advanced treatment options performed even in patients suffering from end-stage lung disease. Due to the positive results of medical informatics in other fields of medicine, lung transplant researchers have also conducted remarkable studies to improve transplant outcomes. The main objective of this article was to review the current studies of health information technology used in lung transplantation.

METHODS

A systematic search was performed in four scientific databases (Web of Science, Scopus, Science Direct, and PubMed) from January 2000 to December 2018. The criteria for inclusion were included in any study describing the use of health information technology or medical informatics in terms of lung transplantation, English papers, and original researchers. The retrieved articles were accordingly screened based on the inclusion and exclusion criteria to select relevant studies. The survey and synthesis of included articles were conducted based on predefined classification.

RESULTS

Out of 263 articles, 27 studies met our inclusion criteria. All included studies involved the application of health information technology in lung transplantation. The types of health information technology methods applied in reviewed articles included mhealth (11.1 %), DSS (7.4 %), decision aid tools (7.4 %), telemedicine (22.2 %), AI methods (11.1 %), data mining (37 %), and patient education (3.7 %). The majority of studies (88.9 %) showed the positive impact of health information technology to enhance lung transplantation outcomes. Finally, the main approaches in different phases of lung transplantation processes were interpreted and summarized in the visual model.

CONCLUSION

This systematic review provides new insights regarding the application of medical informatics in the lung transplantation domain. The missing areas of medical informatics in the lung transplantation domain were recognized through this study.

Primary care physician experiences utilizing a family health history tool with electronic health record-integrated clinical decision support: an implementation process assessment

Family health history (FHH) screening plays a key role in disease risk identification and tailored disease prevention strategies. Primary care physicians (PCPs) are in a frontline position to provide personalized medicine recommendations identified through FHH screening; however, adoption of FHH screening tools has been slow and inconsistent in practice. Information is also lacking on PCP facilitators and barriers of utilizing family history tools with clinical decision support (CDS) embedded in the electronic health record (EHR). This study reports on PCPs' initial experiences with the Genetic and Wellness Assessment (GWA), a patient-administered FHH screening tool utilizing the EHR and CDS. Semi-structured interviews were conducted with 24 PCPs who use the GWA in a network of community-based practices. Four main themes regarding GWA implementation emerged: benefits to clinical care, challenges in practice, CDS-specific issues, and physician-recommended improvements. Sub-themes included value in improving patient access to genetic services, inadequate time to discuss GWA recommendations, lack of patient follow-through with recommendations, and alert fatigue. While PCPs valued the GWA's clinical utility, a number of challenges were identified in the administration and use of the GWA in practice. Based on participants' recommendations, iterative changes have been made to the GWA and workflow to increase efficiency, upgrade the CDS process, and provide additional education to PCPs and patients. Future studies are needed to assess a diverse sample of physicians' and patients' perspectives on the utility of FHH screening utilizing EHR-based genomics recommendations.

Clinical Genome Data Model (cGDM) provides Interactive Clinical Decision Support for Precision Medicine

In light of recent developments in genomic technology and the rapid accumulation of genomic information, a major transition toward precision medicine is anticipated. However, the clinical applications of genomic information remain limited. This lag can be attributed to several complex factors, including the knowledge gap between medical experts and bioinformaticians, the distance between bioinformatics workflows and clinical practice, and the unique characteristics of genomic data, which can make interpretation difficult. Here we present a novel genomic data model that allows for more interactive support in clinical decision-making. Informational modelling was used as a basis to design a communication scheme between sophisticated bioinformatics predictions and the representative data relevant to a clinical decision. This study was conducted by a multidisciplinary working group who carried out clinico-genomic workflow analysis and attribute extraction, through Failure Mode and Effects Analysis (FMEA). Based on those results, a clinical genome data model (cGDM) was developed with 8 entities and 46 attributes. The cGDM integrates reliability-related factors that enable clinicians to access the reliability problem of each individual genetic test result as clinical evidence. The proposed cGDM provides a data-layer infrastructure supporting the intellectual interplay between medical experts and informed decision-making.

Utility of Centrifugation-Controlled Convective (C3) Flow for Rapid On-chip ELISA

Miniaturizing the enzyme-linked immunosorbent assay (ELISA) protocols in microfluidics is sought after by researchers for a rapid, high throughput screening, on-site diagnosis, and ease in operation for detection and quantification of biomarkers. Herein, we report the use of the centrifugation-controlled convective (C3) flow as an alternative method in fluid flow control in a ring-structured channel for enhanced on-chip ELISA. A system that consists of a rotating heater stage and a microfluidic disk chip has been developed and demonstrated to detect IgA. The ring-structured channel was partially filled with microbeads (250 µm in diameter) carrying the capture antibodies and the analyte solution was driven by thermal convection flow (50 µL/min) to promote the reaction. The remaining part of the circular channel without microbeads served as the observation area to measure the absorbance value of the labeled protein. Currently, the system is capable of conducting four reactions in parallel and can be performed within 30 min at 300 G. A detection limit of 6.16 ng/mL using 24 µL of target sample (IgA) was observed. By simply changing the capture antibodies, the system is expected to be versatile for other immunoassays.

Parents', Health Care Professionals', and Scientists' Experiences of a Precision Medicine Pilot Trial for Patients With High-Risk Childhood Cancer: A Qualitative Study

PURPOSE

Children with high-risk cancers have low survival rates because current treatment options are limited. Precision medicine trials are designed to offer patients individualized treatment recommendations, potentially improving their clinical outcomes. However, parents' understanding is often limited, and expectations of benefit to their own child can be high. Health care professionals (HCPs) are often not familiar with precision medicine and might find managing families' expectations challenging. Scientists find themselves working with high expectations among different stakeholders to rapidly translate their identification of actionable targets in real time. Therefore, we wanted to gain an in-depth understanding of the experiences of all stakeholders involved in a new precision medicine pilot trial called TARGET, including parents, their child's HCPs, and the scientists who conducted the laboratory research and generated the data used to make treatment recommendations.

METHODS

We conducted semistructured interviews with all participants and analyzed the interviews thematically.

RESULTS

We interviewed 15 parents (9 mothers; 66.7% bereaved), 17 HCPs, and 16 scientists. We identified the following themes in parents' interviews: minimal understanding and need for more information, hope as a driver of participation, challenges around biopsies, timing, and drug access, and few regrets. HCP and scientist interviews revealed themes such as embracing new technologies and collaborations and challenges managing families' expectations, timing of testing and test results, and drug access.

CONCLUSION

Educating families, HCPs, and scientists to better understand the benefits and limitations of precision medicine trials may improve the transparency of the translation of discovery genomics to novel therapies, increase satisfaction with the child's care, and ameliorate the additional long-term psychosocial burden for families already affected by high-risk childhood cancer.

Automatic cataract grading methods based on deep learning

BACKGROUND AND OBJECTIVE

The shortage of ophthalmologists in rural areas in China causes a lot of cataract patients not getting timely diagnosis and effective treatment. We develop an algorithm and platform to automatically diagnose and grade cataract based on fundus images of patients. This method can help government assisting poor population more accurately.

METHODS

The novel six-level cataract grading method proposed in this paper focuses on the multi-feature fusion based on stacking. We extract two kinds of features which can effectively distinguish different levels of cataract. One is high-level features extracted from residual network (ResNet18). The other is texture features extarcted by gray level co-occurrence matrix (GLCM). Then a frame is proposed to automatically grade cataract by the extracted features. In the frame, two support vector machine (SVM) classifiers are used as base-learners to obtain the probability outputs of each fundus image, and fully connected neural network (FCNN) are used as meta-learner to output the final classification result, which consists of two fully-connected layers.

RESULT

The accuracy of six-level grading achieved by the proposed method is up to 92.66% on average, the highest of which reaches 93.33%. The proposed method achieves 94.75% accuracy on four-level grading for cataract, which is at least 1.75% higher than those of the exiting methods.

CONCLUSIONS

Six-category cataract classification algorithm show that Multi-feature & Stacking proposed in this paper helps achieve higher grading performance and lower volatility than grading using high-level features and texture features respectively. We also apply our algorithm into four-level cataract grading system and it shows higher accuracy compared with previous reports.

Artificial intelligence supported patient self-care in chronic heart failure: a paradigm shift from reactive to predictive, preventive and personalised care

Heart failure (HF) is one of the most complex chronic disorders with high prevalence, mainly due to the ageing population and better treatment of underlying diseases. Prevalence will continue to rise and is estimated to reach 3% of the population in Western countries by 2025. It is the most important cause of hospitalisation in subjects aged 65 years or more, resulting in high costs and major social impact. The current "one-size-fits-all" approach in the treatment of HF does not result in best outcome for all patients. These facts are an imminent threat to good quality management of patients with HF. An unorthodox approach from a new vision on care is required. We propose a novel predictive, preventive and personalised medicine approach where patients are truly leading their management, supported by an easily accessible online application that takes advantage of artificial intelligence. This strategy paper describes the needs in HF care, the needed paradigm shift and the elements that are required to achieve this shift. Through the inspiring collaboration of clinical and high-tech partners from North-West Europe combining state of the art HF care, artificial intelligence, serious gaming and patient coaching, a virtual doctor is being created. The results are expected to advance and personalise self-care, where standard care tasks are performed by the patients themselves, in principle without involvement of healthcare professionals, the latter being able to focus on complex conditions. This new vision on care will significantly reduce costs per patient while improving outcomes to enable long-term sustainability of top-level HF care.

Structured Data Acquisition in Oncology

BACKGROUND

Demographic changes and improvement in therapy have shifted the focus of treatment towards chronic diseases and multiple health conditions. This has caused a tremendous increase in data per patient that needs to be integrated longitudinally and across departmental silos. The general increase in the volume of data per diagnostic examination and the number of diagnostic procedures per diagnostic pathway additionally accentuate this data integration challenge.

SUMMARY

Subspecialization in medicine has led to largely autonomously organized departments with in-dependent IT ecosystems. This patchwork of IT infrastructure is not prepared to meet the data integration challenge. The resulting lack of integrated information makes the treatment of chronically ill patients increasingly difficult and error prone. Key Message: A sustainable method for data ac-quisition is needed to aid multimodal treatment and improve efficiency in healthcare.

Merits, features, and desiderata to be considered when developing electronic health records with embedded clinical decision support systems in Palestinian hospitals: a consensus study

BACKGROUND

Electronic health records (EHRs) with embedded clinical decision support systems (CDSSs) have the potential to improve healthcare delivery. This study was conducted to explore merits, features, and desiderata to be considered when planning for, designing, developing, implementing, piloting, evaluating, maintaining, upgrading, and/or using EHRs with CDSSs.

METHODS

A mixed-method combining the Delphi technique and Analytic Hierarchy Process was used. Potentially important items were collected after a thorough search of the literature and from interviews with key contact experts (n = 19). Opinions and views of the 76 panelists on the use of EHRs were also explored. Iterative Delphi rounds were conducted to achieve consensus on 122 potentially important items by a panel of 76 participants. Items on which consensus was achieved were ranked in the order of their importance using the Analytic Hierarchy Process.

RESULTS

Of the 122 potentially important items presented to the panelists in the Delphi rounds, consensus was achieved on 110 (90.2%) items. Of these, 16 (14.5%) items were related to the demographic characteristics of the patient, 16 (14.5%) were related to prescribing medications, 16 (14.5%) were related to checking prescriptions and alerts, 14 (12.7%) items were related to the patient's identity, 13 (11.8%) items were related to patient assessment, 12 (10.9%) items were related to the quality of alerts, 11 (10%) items were related to admission and discharge of the patient, 9 (8.2%) items were general features, and 3 (2.7%) items were related to diseases and making diagnosis.

CONCLUSIONS

In this study, merits, features, and desiderata to be considered when planning for, designing, developing, implementing, piloting, evaluating, maintaining, upgrading, and/or using EHRs with CDSSs were explored. Considering items on which consensus was achieved might promote congruence and safe use of EHRs. Further studies are still needed to determine if these recommendations can improve patient safety and outcomes in Palestinian hospitals.

An Optimisation-Driven Prediction Method for Automated Diagnosis and Prognosis

This article presents a novel hybrid classification paradigm for medical diagnoses and prognoses prediction. The core mechanism of the proposed method relies on a centroid classification algorithm whose logic is exploited to formulate the classification task as a real-valued optimisation problem. A novel metaheuristic combining the algorithmic structure of Swarm Intelligence optimisers with the probabilistic search models of Estimation of Distribution Algorithms is designed to optimise such a problem, thus leading to high-accuracy predictions. This method is tested over 11 medical datasets and compared against 14 cherry-picked classification algorithms. Results show that the proposed approach is competitive and superior to the state-of-the-art on several occasions.

Developing a framework for evidence-based grading and assessment of predictive tools for clinical decision support

BACKGROUND

Clinical predictive tools quantify contributions of relevant patient characteristics to derive likelihood of diseases or predict clinical outcomes. When selecting predictive tools for implementation at clinical practice or for recommendation in clinical guidelines, clinicians are challenged with an overwhelming and ever-growing number of tools, most of which have never been implemented or assessed for comparative effectiveness. To overcome this challenge, we have developed a conceptual framework to Grade and Assess Predictive tools (GRASP) that can provide clinicians with a standardised, evidence-based system to support their search for and selection of efficient tools.

METHODS

A focused review of the literature was conducted to extract criteria along which tools should be evaluated. An initial framework was designed and applied to assess and grade five tools: LACE Index, Centor Score, Well's Criteria, Modified Early Warning Score, and Ottawa knee rule. After peer review, by six expert clinicians and healthcare researchers, the framework and the grading of the tools were updated.

RESULTS

GRASP framework grades predictive tools based on published evidence across three dimensions: 1) Phase of evaluation; 2) Level of evidence; and 3) Direction of evidence. The final grade of a tool is based on the highest phase of evaluation, supported by the highest level of positive evidence, or mixed evidence that supports a positive conclusion. Ottawa knee rule had the highest grade since it has demonstrated positive post-implementation impact on healthcare. LACE Index had the lowest grade, having demonstrated only pre-implementation positive predictive performance.

CONCLUSION

GRASP framework builds on widely accepted concepts to provide standardised assessment and evidence-based grading of predictive tools. Unlike other methods, GRASP is based on the critical appraisal of published evidence reporting the tools' predictive performance before implementation, potential effect and usability during implementation, and their post-implementation impact. Implementing the GRASP framework as an online platform can enable clinicians and guideline developers to access standardised and structured reported evidence of existing predictive tools. However, keeping GRASP reports up-to-date would require updating tools' assessments and grades when new evidence becomes available, which can only be done efficiently by employing semi-automated methods for searching and processing the incoming information.

A practical computerized decision support system for predicting the severity of Alzheimer's disease of an individual

Computerized clinical decision support systems can help to provide objective, standardized, and timely dementia diagnosis. However, current computerized systems are mainly based on group analysis, discrete classification of disease stages, or expensive and not readily accessible biomarkers, while current clinical practice relies relatively heavily on cognitive and functional assessments (CFA). In this study, we developed a computational framework using a suite of machine learning tools for identifying key markers in predicting the severity of Alzheimer's disease (AD) from a large set of biological and clinical measures. Six machine learning approaches, namely Kernel Ridge Regression (KRR), Support Vector Regression, and k-Nearest Neighbor for regression and Support Vector Machine (SVM), Random Forest, and k-Nearest Neighbor for classification, were used for the development of predictive models. We demonstrated high predictive power of CFA. Predictive performance of models incorporating CFA was shown to consistently have higher accuracy than those based solely on biomarker modalities. We found that KRR and SVM were the best performing regression and classification methods respectively. The optimal SVM performance was observed for a set of four CFA test scores (FAQ, ADAS13, MoCA, MMSE) with multi-class classification accuracy of 83.0%, 95%CI = (72.1%, 93.8%) while the best performance of the KRR model was reported with combined CFA and MRI neuroimaging data, i.e., R ² = 0.874, 95%CI = (0.827, 0.922). Given the high predictive power of CFA and their widespread use in clinical practice, we then designed a data-driven and self-adaptive computerized clinical decision support system (CDSS) prototype for evaluating the severity of AD of an individual on a continuous spectrum. The system implemented an automated computational approach for data pre-processing, modelling, and validation and used exclusively the scores of selected cognitive measures as data entries. Taken together, we have developed an objective and practical CDSS to aid AD diagnosis.

Diffusion dynamics of electronic health records: A longitudinal observational study comparing data from hospitals in Germany and the United States

BACKGROUND

While aiming for the same goal of building a national eHealth Infrastructure, Germany and the United States pursued different strategic approaches - particularly regarding the role of promoting the adoption and usage of hospital Electronic Health Records (EHR).

OBJECTIVE

To measure and model the diffusion dynamics of EHRs in German hospital care and to contrast the results with the developments in the US.

MATERIALS AND METHODS

All acute care hospitals that were members of the German statutory health system were surveyed during the period 2007-2017 for EHR adoption. Bass models were computed based on the German data and the corresponding data of the American Hospital Association (AHA) from non-federal hospitals in order to model and explain the diffusion of innovation.

RESULTS

While the diffusion dynamics observed in the US resembled the typical s-shaped curve with high imitation effects (q = 0.583) but with a relatively low innovation effect (p = 0.025), EHR diffusion in Germany stagnated with adoption rates of approx. 50% (imitation effect q = -0.544) despite a higher innovation effect (p = 0.303).

DISCUSSION

These findings correlate with different governmental strategies in the US and Germany of financially supporting EHR adoption. Imitation only seems to work if there are financial incentives, e.g. those of the HITECH Act in the US. They are lacking in Germany, where the government left health IT adoption strategies solely to the free market and the consensus among all of the stakeholders.

CONCLUSION

Bass diffusion models proved to be useful for distinguishing the diffusion dynamics in German and US non-federal hospitals. When applying the Bass model, the imitation parameter needs a broader interpretation beyond the network effects, including driving forces such as incentives and regulations, as was demonstrated by this study.

Machine learning-based coronary artery disease diagnosis: A comprehensive review

Coronary artery disease (CAD) is the most common cardiovascular disease (CVD) and often leads to a heart attack. It annually causes millions of deaths and billions of dollars in financial losses worldwide. Angiography, which is invasive and risky, is the standard procedure for diagnosing CAD. Alternatively, machine learning (ML) techniques have been widely used in the literature as fast, affordable, and noninvasive approaches for CAD detection. The results that have been published on ML-based CAD diagnosis differ substantially in terms of the analyzed datasets, sample sizes, features, location of data collection, performance metrics, and applied ML techniques. Due to these fundamental differences, achievements in the literature cannot be generalized. This paper conducts a comprehensive and multifaceted review of all relevant studies that were published between 1992 and 2019 for ML-based CAD diagnosis. The impacts of various factors, such as dataset characteristics (geographical location, sample size, features, and the stenosis of each coronary artery) and applied ML techniques (feature selection, performance metrics, and method) are investigated in detail. Finally, the important challenges and shortcomings of ML-based CAD diagnosis are discussed.

CADDIE2-evaluation of a clinical decision-support system for early detection of systemic inflammatory response syndrome in paediatric intensive care: study protocol for a diagnostic study

INTRODUCTION

Systemic inflammatory response syndrome (SIRS) is one of the most critical indicators determining the clinical outcome of paediatric intensive care patients. Clinical decision support systems (CDSS) can be designed to support clinicians in detection and treatment. However, the use of such systems is highly discussed as they are often associated with accuracy problems and 'alert fatigue'. We designed a CDSS for detection of paediatric SIRS and hypothesise that a high diagnostic accuracy together with an adequate alerting will accelerate the use. Our study will (1) determine the diagnostic accuracy of the CDSS compared with gold standard decisions created by two blinded, experienced paediatricians, and (2) compare the system's diagnostic accuracy with that of routine clinical care decisions compared with the same gold standard.

METHODS AND ANALYSIS

CADDIE2 is a prospective diagnostic accuracy study taking place at the Department of Pediatric Cardiology and Intensive Care Medicine at the Hannover Medical School; it represents the second step towards our vision of cross-institutional and data-driven decision-support for intensive care environments (CADDIE). The study comprises (1) recruitment of up to 300 patients (start date 1 August 2018), (2) creation of gold standard decisions (start date 1 May 2019), (3) routine SIRS assessments by physicians (starts with recruitment), (4) SIRS assessments by a CDSS (start date 1 May 2019), and (5) statistical analysis with a modified approach for determining sensitivity and specificity and comparing the accuracy results of the different diagnostic approaches (planned start date 1 July 2019).

ETHICS AND DISSEMINATION

Ethics approval was obtained at the study centre (Ethics Committee of Hannover Medical School). Results of the main study will be communicated via publication in a peer-reviewed journal.

TRIAL REGISTRATION NUMBER

ClinicalTrials.gov NCT03661450; Pre-results.

Physician Review of a Celiac Disease Risk Estimation and Decision-Making Expert System

Objective: Celiac disease is a genetic disease affecting people of all ages, resulting in small intestine enteropathy. It is considered to be a clinical chameleon. Average prevalence of celiac disease is 1 out of 100 people with data indicating the risk may be as high as 22% for those with first-degree relatives with the disease. Eighty-three percent of people with celiac disease may be undiagnosed. Average duration to diagnosis is 10 years. Data indicate that there is a lack of consensus regarding diagnostics and symptomatology.Method: A clinical decision support system (CDSS) was developed using Exsys Corvid for expert analysis (CD-CDSS). The CD-CDSS was divided into symptoms and manifestations with 80 points of navigation, and a serology section, and was validated by 13 experts in the field of celiac disease using a 10-statement 5-point Likert scale.Results: This scale was analyzed using Cronbach's alpha reliability coefficient, which was calculated using SPSS and revealed good internal consistency and reliability with a result of 0.813. One hundred percent of experts agreed that the CD-CDSS is capable of guiding a health care professional through the diagnostic process, contains an accurate list of symptoms based on the clinical literature, and can foster improved awareness and education about celiac disease and that there is a need for this system.Conclusions: A celiac disease risk estimation and decision-making expert system was successfully developed and evaluated by medical professionals, with 100% agreeing that this CD-CDSS is medically accurate and can guide health care professionals through the diagnostic process.

Technical, Biological, and Systems Barriers for Molecular Clinical Decision Support

Genome-enabled or molecular clinical decision support (CDS) systems provide unique advantages for the clinical use of genomic data; however, their implementation is complicated by technical, biological, and systemic barriers. This article reviews the substantial technical progress that has been made in the past decade and finds that the underlying biological limitations of genomics as well as systemic barriers to adoption of molecular CDS have been comparatively underestimated. A hybrid consultative CDS system, which integrates a genomics consultant into an active CDS system, may provide an interim path forward.

Diagnostic Efficacy and Therapeutic Decision-making Capacity of an Artificial Intelligence Platform for Childhood Cataracts in Eye Clinics: A Multicentre Randomized Controlled Trial

BACKGROUND

CC-Cruiser is an artificial intelligence (AI) platform developed for diagnosing childhood cataracts and providing risk stratification and treatment recommendations. The high accuracy of CC-Cruiser was previously validated using specific datasets. The objective of this study was to compare the diagnostic efficacy and treatment decision-making capacity between CC-Cruiser and ophthalmologists in real-world clinical settings.

METHODS

This multicentre randomized controlled trial was performed in five ophthalmic clinics in different areas across China. Pediatric patients (aged ≤ 14 years) without a definitive diagnosis of cataracts or history of previous eye surgery were randomized (1:1) to receive a diagnosis and treatment recommendation from either CC-Cruiser or senior consultants (with over 5 years of clinical experience in pediatric ophthalmology). The experts who provided a gold standard diagnosis, and the investigators who performed slit-lamp photography and data analysis were blinded to the group assignments. The primary outcome was the diagnostic performance for childhood cataracts with reference to cataract experts' standards. The secondary outcomes included the evaluation of disease severity and treatment determination, the time required for the diagnosis, and patient satisfaction, which was determined by the mean rating. This trial is registered with ClinicalTrials.gov (NCT03240848).

FINDINGS

Between August 9, 2017 and May 25, 2018, 350 participants (700 eyes) were randomly assigned for diagnosis by CC-Cruiser (350 eyes) or senior consultants (350 eyes). The accuracies of cataract diagnosis and treatment determination were 87.4% and 70.8%, respectively, for CC-Cruiser, which were significantly lower than 99.1% and 96.7%, respectively, for senior consultants (p < 0.001, OR = 0.06 [95% CI 0.02 to 0.19]; and p < 0.001, OR = 0.08 [95% CI 0.03 to 0.25], respectively). The mean time for receiving a diagnosis from CC-Cruiser was 2.79 min, which was significantly less than 8.53 min for senior consultants (p < 0.001, mean difference 5.74 [95% CI 5.43 to 6.05]). The patients were satisfied with the overall medical service quality provided by CC-Cruiser, typically with its time-saving feature in cataract diagnosis.

INTERPRETATION

CC-Cruiser exhibited less accurate performance comparing to senior consultants in diagnosing childhood cataracts and making treatment decisions. However, the medical service provided by CC-Cruiser was less time-consuming and achieved a high level of patient satisfaction. CC-Cruiser has the capacity to assist human doctors in clinical practice in its current state.

FUNDING

National Key R&D Program of China (2018YFC0116500) and the Key Research Plan for the National Natural Science Foundation of China in Cultivation Project (91846109).

Data-driven clinical decision processes: it's time

Changes and transformations enabled by Big Data have direct effects on Translational Medicine. At one end, superior precision is expected from a more data-intensive and individualized medicine, thus accelerating scientific discovery and innovation (in diagnosis, therapy, disease management etc.). At the other end, the scientific method needs to adapt to the increased diversity that data present, and this can be beneficial because potentially revealing greater details of how a disease manifests and progresses. Patient-focused health data provides augmented complexity too, far beyond the simple need of testing hypotheses or validating models. Clinical decision support systems (CDSS) will increasingly deal with such complexity by developing efficient high-performance algorithms and creating a next generation of inferential tools for clinical use. Additionally, new protocols for sharing digital information and effectively integrating patients data will need to be CDSS-embedded features in view of suitable data harmonization aimed at improved diagnosis, therapy assessment and prevention.

Applications of Blockchain Technology in Medicine and Healthcare: Challenges and Future Perspectives

Blockchain technology has gained considerable attention, with an escalating interest in a plethora of numerous applications, ranging from data management, financial services, cyber security, IoT, and food science to healthcare industry and brain research. There has been a remarkable interest witnessed in utilizing applications of blockchain for the delivery of safe and secure healthcare data management. Also, blockchain is reforming the traditional healthcare practices to a more reliable means, in terms of effective diagnosis and treatment through safe and secure data sharing. In the future, blockchain could be a technology that may potentially help in personalized, authentic, and secure healthcare by merging the entire real-time clinical data of a patient’s health and presenting it in an up-to-date secure healthcare setup. In this paper, we review both the existing and latest developments in the field of healthcare by implementing blockchain as a model. We also discuss the applications of blockchain, along with the challenges faced and future perspectives.

CDSS-RM: a clinical decision support system reference model

Clinical Decision Support Systems (CDSS) provide aid in clinical decision making and therefore need to take into consideration human, data interactions, and cognitive functions of clinical decision makers. The objective of this paper is to introduce a high level reference model that is intended to be used as a foundation to design successful and contextually relevant CDSS systems. The paper begins by introducing the information flow, use, and sharing characteristics in a hospital setting, and then it outlines the referential context for the model, which are clinical decisions in a hospital setting. Important characteristics of the Clinical decision making process include: (i) Temporally ordered steps, each leading to new data, which in turn becomes useful for a new decision, (ii) Feedback loops where acquisition of new data improves certainty and generates new questions to examine, (iii) Combining different kinds of clinical data for decision making, (iv) Reusing the same data in two or more different decisions, and (v) Clinical decisions requiring human cognitive skills and knowledge, to process the available information. These characteristics form the foundation to delineate important considerations of Clinical Decision Support Systems design. The model includes six interacting and interconnected elements, which formulate the high-level reference model (CDSS-RM). These elements are introduced in the form of questions, as considerations, and are examined with the use of illustrated scenario-based and data-driven examples. The six elements /considerations of the reference model are: (i) Do CDSS mimic the cognitive process of clinical decision makers? (ii) Do CDSS provide recommendations with longitudinal insight? (iii) Is the model performance contextually realistic? (iv) Is the ‘Historical Decision’ bias taken into consideration in CDSS design? (v) Do CDSS integrate established clinical standards and protocols? (vi) Do CDSS utilize unstructured data? The CDSS-RM reference model can contribute to optimized design of modeling methodologies, in order to improve response of health systems to clinical decision-making challenges.

Applying Artificial Intelligence to Address the Knowledge Gaps in Cancer Care

BACKGROUND

Rapid advances in science challenge the timely adoption of evidence-based care in community settings. To bridge the gap between what is possible and what is practiced, we researched approaches to developing an artificial intelligence (AI) application that can provide real-time patient-specific decision support.

MATERIALS AND METHODS

The Oncology Expert Advisor (OEA) was designed to simulate peer-to-peer consultation with three core functions: patient history summarization, treatment options recommendation, and management advisory. Machine-learning algorithms were trained to construct a dynamic summary of patients cancer history and to suggest approved therapy or investigative trial options. All patient data used were retrospectively accrued. Ground truth was established for approximately 1,000 unique patients. The full Medline database of more than 23 million published abstracts was used as the literature corpus.

RESULTS

OEA's accuracies of searching disparate sources within electronic medical records to extract complex clinical concepts from unstructured text documents varied, with F1 scores of 90%-96% for non-time-dependent concepts (e.g., diagnosis) and F1 scores of 63%-65% for time-dependent concepts (e.g., therapy history timeline). Based on constructed patient profiles, OEA suggests approved therapy options linked to supporting evidence (99.9% recall; 88% precision), and screens for eligible clinical trials on ClinicalTrials.gov (97.9% recall; 96.9% precision).

CONCLUSION

Our results demonstrated technical feasibility of an AI-powered application to construct longitudinal patient profiles in context and to suggest evidence-based treatment and trial options. Our experience highlighted the necessity of collaboration across clinical and AI domains, and the requirement of clinical expertise throughout the process, from design to training to testing.

IMPLICATIONS FOR PRACTICE

Artificial intelligence (AI)-powered digital advisors such as the Oncology Expert Advisor have the potential to augment the capacity and update the knowledge base of practicing oncologists. By constructing dynamic patient profiles from disparate data sources and organizing and vetting vast literature for relevance to a specific patient, such AI applications could empower oncologists to consider all therapy options based on the latest scientific evidence for their patients, and help them spend less time on information "hunting and gathering" and more time with the patients. However, realization of this will require not only AI technology maturation but also active participation and leadership by clincial experts.

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.