Rapid Learning Health Care Research Articles

Confederated learning in healthcare: Training machine learning models using disconnected data separated by individual, data type and identity for Large-Scale health system Intelligence

BackgroundA patient’s health information is generally fragmented across silos because it follows how care is delivered: multiple providers in multiple settings. Though it is technically feasible to reunite data for analysis in a manner that underpins a rapid learning healthcare system, privacy concerns and regulatory barriers limit data centralization for this purpose. ObjectivesMachine learning can be conducted in a federated manner on patient datasets with the same set of variables but separated across storage. But federated learning cannot handle the situation where different data types for a given patient are separated vertically across different organizations and when patient ID matching across different institutions is difficult. We call methods that enable machine learning model training on data separated by two or more dimensions “confederated machine learning”, which we aim to develop in this study. MethodsWe propose and evaluate confederated learning for training machine learning models to stratify the risk of several diseases among silos when data are horizontally separated by individual, vertically separated by data type, and separated by identity without patient ID matching. The confederated learning method can be intuitively understood as a distributed learning method with representation learning, generative model, imputation method and data augmentation elements. ResultsOur confederated learning method achieves AUCROC (Area Under The Curve Receiver Operating Characteristics) of 0.787 for diabetes prediction, 0.718 for psychological disorders prediction, and 0.698 for Ischemic heart disease prediction using nationwide health insurance claims. ConclusionOur proposed confederated learning method successfully trained machine learning models on health insurance data separated by two or more dimensions.

The Assisi Think Tank Meeting Breast Large Database for Standardized Data Collection in Breast Cancer—ATTM.BLADE

Background: During the 2016 Assisi Think Tank Meeting (ATTM) on breast cancer, the panel of experts proposed developing a validated system, based on rapid learning health care (RLHC) principles, to standardize inter-center data collection and promote personalized treatments for breast cancer. Material and Methods: The seven-step Breast LArge DatabasE (BLADE) project included data collection, analysis, application, and evaluation on a data-sharing platform. The multidisciplinary team developed a consensus-based ontology of validated variables with over 80% agreement. This English-language ontology constituted a breast cancer library with seven knowledge domains: baseline, primary systemic therapy, surgery, adjuvant systemic therapies, radiation therapy, follow-up, and toxicity. The library was uploaded to the BLADE domain. The safety of data encryption and preservation was tested according to General Data Protection Regulation (GDPR) guidelines on data from 15 clinical charts. The system was validated on 64 patients who had undergone post-mastectomy radiation therapy. In October 2018, the BLADE system was approved by the Ethical Committee of Fondazione Policlinico Gemelli IRCCS, Rome, Italy (Protocol No. 0043996/18). Results: From June 2016 to July 2019, the multidisciplinary team completed the work plan. An ontology of 218 validated variables was uploaded to the BLADE domain. The GDPR safety test confirmed encryption and data preservation (on 5000 random cases). All validation benchmarks were met. Conclusion: BLADE is a support system for follow-up and assessment of breast cancer care. To successfully develop and validate it as the first standardized data collection system, multidisciplinary collaboration was crucial in selecting its ontology and knowledge domains. BLADE is suitable for multi-center uploading of retrospective and prospective clinical data, as it ensures anonymity and data privacy.

Deep learning for natural language processing of free-text pathology reports: a comparison of learning curves

IntroductionAlthough clinically derived information could improve patient care, its full potential remains unrealised because most of it is stored in a format unsuitable for traditional methods of analysis, free-text clinical...

Assessment of Deep Natural Language Processing in Ascertaining Oncologic Outcomes From Radiology Reports

A rapid learning health care system for oncology will require scalable methods for extracting clinical end points from electronic health records (EHRs). Outside of clinical trials, end points such as cancer progression and response are not routinely encoded into structured data. To determine whether deep natural language processing can extract relevant cancer outcomes from radiologic reports, a ubiquitous but unstructured EHR data source. A retrospective cohort study evaluated 1112 patients who underwent tumor genotyping for a diagnosis of lung cancer and participated in the Dana-Farber Cancer Institute PROFILE study from June 26, 2013, to July 2, 2018. Patients were divided into curation and reserve sets. Human abstractors applied a structured framework to radiologic reports for the curation set to ascertain the presence of cancer and changes in cancer status over time (ie, worsening/progressing vs improving/responding). Deep learning models were then trained to capture these outcomes from report text and subsequently evaluated in a 10% held-out test subset of curation patients. Cox proportional hazards regression models compared human and machine curations of disease-free survival, progression-free survival, and time to improvement/response in the curation set, and measured associations between report classification and overall survival in the curation and reserve sets. The primary outcome was area under the receiver operating characteristic curve (AUC) for deep learning models; secondary outcomes were time to improvement/response, disease-free survival, progression-free survival, and overall survival. A total of 2406 patients were included (mean [SD] age, 66.5 [10.8] years; 1428 female [59.7%]; 2170 [90.2%] white). Radiologic reports (n = 14 230) were manually reviewed for 1112 patients in the curation set. In the test subset (n = 109), deep learning models identified the presence of cancer, improvement/response, and worsening/progression with accurate discrimination (AUC >0.90). Machine and human curation yielded similar measurements of disease-free survival (hazard ratio [HR] for machine vs human curation, 1.18; 95% CI, 0.71-1.95); progression-free survival (HR, 1.11; 95% CI, 0.71-1.71); and time to improvement/response (HR, 1.03; 95% CI, 0.65-1.64). Among 15 000 additional reports for 1294 reserve set patients, algorithm-detected cancer worsening/progression was associated with decreased overall survival (HR for mortality, 4.04; 95% CI, 2.78-5.85), and improvement/response was associated with increased overall survival (HR, 0.41; 95% CI, 0.22-0.77). Deep natural language processing appears to speed curation of relevant cancer outcomes and facilitate rapid learning from EHR data.

How Big Data, Comparative Effectiveness Research, and Rapid-Learning Health-Care Systems Can Transform Patient Care in Radiation Oncology.

Big data and comparative effectiveness research methodologies can be applied within the framework of a rapid-learning health-care system (RLHCS) to accelerate discovery and to help turn the dream of fully personalized medicine into a reality. We synthesize recent advances in genomics with trends in big data to provide a forward-looking perspective on the potential of new advances to usher in an era of personalized radiation therapy, with emphases on the power of RLHCS to accelerate discovery and the future of individualized radiation treatment planning.

Infrastructure and distributed learning methodology for privacy-preserving multi-centric rapid learning health care: euroCAT

Machine learning applications for personalized medicine are highly dependent on access to sufficient data. For personalized radiation oncology, datasets representing the variation in the entire cancer patient population need to be acquired and used to learn prediction models. Ethical and legal boundaries to ensure data privacy hamper collaboration between research institutes. We hypothesize that data sharing is possible without identifiable patient data leaving the radiation clinics and that building machine learning applications on distributed datasets is feasible.We developed and implemented an IT infrastructure in five radiation clinics across three countries (Belgium, Germany, and The Netherlands). We present here a proof-of-principle for future ‘big data’ infrastructures and distributed learning studies. Lung cancer patient data was collected in all five locations and stored in local databases. Exemplary support vector machine (SVM) models were learned using the Alternating Direction Method of Multipliers (ADMM) from the distributed databases to predict post-radiotherapy dyspnea grade ⩾2. The discriminative performance was assessed by the area under the curve (AUC) in a five-fold cross-validation (learning on four sites and validating on the fifth). The performance of the distributed learning algorithm was compared to centralized learning where datasets of all institutes are jointly analyzed.The euroCAT infrastructure has been successfully implemented in five radiation clinics across three countries. SVM models can be learned on data distributed over all five clinics. Furthermore, the infrastructure provides a general framework to execute learning algorithms on distributed data. The ongoing expansion of the euroCAT network will facilitate machine learning in radiation oncology. The resulting access to larger datasets with sufficient variation will pave the way for generalizable prediction models and personalized medicine.

Building a Rapid Learning Health Care System for Oncology: Why CancerLinQ Collects Identifiable Health Information to Achieve Its Vision.

The ever-increasing volume of scientific discoveries, clinical knowledge, novel diagnostic tools, and treatment options juxtaposed with rising costs in health care challenge physicians to identify, prioritize, and use new information rapidly to deliver efficient and high-quality care to a growing and aging patient population. CancerLinQ, a rapid learning health care system in oncology, is an initiative of the American Society of Clinical Oncology and its Institute for Quality that addresses these challenges by collecting information from the electronic health records of large numbers of patients with cancer. CancerLinQ is first and foremost a quality measurement and reporting system through which oncologists can harness the depth and power of their patients' clinical records and other data to assess, monitor, and improve the care they deliver. However, in light of privacy and security concerns with regard to collection, use, and disclosure of patient information, this article addresses the need to collect protected health information as defined under the Health Insurance Portability and Accountability Act of 1996 to drive rapid learning through CancerLinQ.

The VA Point-of-Care Precision Oncology Program: Balancing Access with Rapid Learning in Molecular Cancer Medicine.

The Department of Veterans Affairs (VA) recognized the need to balance patient-centered care with responsible creation of generalizable knowledge on the effectiveness of molecular medicine tools. Embracing the principles of the rapid learning health-care system, a new clinical program called the Precision Oncology Program (POP) was created in New England. The POP integrates generalized knowledge about molecular medicine in cancer with a database of observations from previously treated veterans. The program assures access to modern genomic oncology practice in the veterans affairs (VA), removes disparities of access across the VA network of clinical centers, disseminates the products of learning that are generalizable to non-VA settings, and systematically presents opportunities for patients to participate in clinical trials of targeted therapeutics.

Standardized data collection to build prediction models in oncology: a prototype for rectal cancer.

The advances in diagnostic and treatment technology are responsible for a remarkable transformation in the internal medicine concept with the establishment of a new idea of personalized medicine. Inter- and intra-patient tumor heterogeneity and the clinical outcome and/or treatment's toxicity's complexity, justify the effort to develop predictive models from decision support systems. However, the number of evaluated variables coming from multiple disciplines: oncology, computer science, bioinformatics, statistics, genomics, imaging, among others could be very large thus making traditional statistical analysis difficult to exploit. Automated data-mining processes and machine learning approaches can be a solution to organize the massive amount of data, trying to unravel important interaction. The purpose of this paper is to describe the strategy to collect and analyze data properly for decision support and introduce the concept of an 'umbrella protocol' within the framework of 'rapid learning healthcare'.

Connecting the Dots: Linking the National Program of Cancer Registries and the Needs of Survivors and Clinicians.

Cancer survivors, the medical community, public health professionals, researchers, and policymakers all need information about newly diagnosed cancer cases and deaths to better understand and address the disease burden. CDC collects cancer data on 96% of the U.S. population through the National Program of Cancer Registries. The National Program of Cancer Registries routinely collects data on all cancer occurrences, deaths, and the types of initial treatment received by the patients, and recently CDC has made advances in its cancer surveillance activities that have direct applicability to cancer survivorship research and care. This article examines CDC's innovative uses of the National Program of Cancer Registries infrastructure and data as a recruitment source for survivorship research studies and behavioral interventions; comparative effectiveness and patient-centered outcomes research; and the collection, consolidation, and dissemination of treatment summaries for cancer survivors and their providers. This paper also discusses long-term, idealistic plans for additional data linkages and sharing among public health, providers, and the cancer survivor through innovative concepts such as patient portals and rapid-learning health care.

Modern clinical research: How rapid learning health care and cohort multiple randomised clinical trials complement traditional evidence based medicine

ABSTRACTBackground. Trials are vital in informing routine clinical care; however, current designs have major deficiencies. An overview of the various challenges that face modern clinical research and the methods that can be exploited to solve these challenges, in the context of personalised cancer treatment in the 21st century is provided.Aim. The purpose of this manuscript, without intending to be comprehensive, is to spark thought whilst presenting and discussing two important and complementary alternatives to traditional evidence-based medicine, specifically rapid learning health care and cohort multiple randomised controlled trial design. Rapid learning health care is an approach that proposes to extract and apply knowledge from routine clinical care data rather than exclusively depending on clinical trial evidence, (please watch the animation: http://youtu.be/ZDJFOxpwqEA). The cohort multiple randomised controlled trial design is a pragmatic method which has been proposed to help overcome the weaknesses of conventional randomised trials, taking advantage of the standardised follow-up approaches more and more used in routine patient care. This approach is particularly useful when the new intervention is a priori attractive for the patient (i.e. proton therapy, patient decision aids or expensive medications), when the outcomes are easily collected, and when there is no need of a placebo arm.Discussion. Truly personalised cancer treatment is the goal in modern radiotherapy. However, personalised cancer treatment is also an immense challenge. The vast variety of both cancer patients and treatment options makes it extremely difficult to determine which decisions are optimal for the individual patient. Nevertheless, rapid learning health care and cohort multiple randomised controlled trial design are two approaches (among others) that can help meet this challenge.

PO-0903: Improving prediction models in the era of rapid learning health care: weighting data to reflect relative importance

PO-0903: Improving prediction models in the era of rapid learning health care: weighting data to reflect relative importance

Building a rapid learning health care system for oncology: the regulatory framework of CancerLinQ.

Today is a time of unprecedented opportunity and challenge in health care generally and cancer care in particular. An explosion of scientific knowledge, the rapid introduction of new drugs and technologies, and the unprecedented escalation in the cost of health care challenge physicians to quickly assimilate new information and appropriately deploy new advances while also delivering efficient and high-quality care to a rapidly growing and aging patient population. At the same time, big data, with its potential to drive rapid understanding and innovation, promises to transform the health care industry, as it has many others already. CancerLinQ is an initiative of the American Society of Clinical Oncology (ASCO) and its Institute for Quality, developed to build on these opportunities and address these challenges by collecting information from the electronic health records of large numbers of patients with cancer. CancerLinQ is, first and foremost, a quality measurement and reporting system through which oncologists can harness the depth and power of their patients' clinical records and other data to improve the care they deliver. The development and deployment of CancerLinQ raises many important questions about the use of big data in health care. This article focuses on the US federal regulatory pathway by which CancerLinQ will accept patient records and the approach of ASCO toward stewardship of the information.

Patient-centred interprofessional collaboration in primary care: challenges for clinical, educational and health services research. An EGPRN keynote paper

The theme ‘patient-centred interprofessional collaboration’ of the EGPRN conference in October 2012, captures in just three words important challenges for European primary care and its research agenda. Challenges for future research are formulated, in three domains: clinical, educational and health services research. Transferability of research, based upon advanced computational infrastructure, will facilitate a rapid learning health care system. In educational research, this includes the use of observational and reflexivity methods. Outcomes should be defined in terms of improvement of functional status and social participation rather than in terms of disease-specific outcomes. Partnership with all stakeholders, patients, GPs and their health care colleagues and students, can help in reducing avoidable waste in the production and reporting of research evidence.

Collaborative Biomedicine in the Age of Big Data: The Case of Cancer

Biomedicine is undergoing a revolution driven by high throughput and connective computing that is transforming medical research and practice. Using oncology as an example, the speed and capacity of genomic sequencing technologies is advancing the utility of individual genetic profiles for anticipating risk and targeting therapeutics. The goal is to enable an era of “P4” medicine that will become increasingly more predictive, personalized, preemptive, and participative over time. This vision hinges on leveraging potentially innovative and disruptive technologies in medicine to accelerate discovery and to reorient clinical practice for patient-centered care. Based on a panel discussion at the Medicine 2.0 conference in Boston with representatives from the National Cancer Institute, Moffitt Cancer Center, and Stanford University School of Medicine, this paper explores how emerging sociotechnical frameworks, informatics platforms, and health-related policy can be used to encourage data liquidity and innovation. This builds on the Institute of Medicine’s vision for a “rapid learning health care system” to enable an open source, population-based approach to cancer prevention and control.

‘Rapid Learning health care in oncology’ – An approach towards decision support systems enabling customised radiotherapy’

PurposeAn overview of the Rapid Learning methodology, its results, and the potential impact on radiotherapy. Material and resultsRapid Learning methodology is divided into four phases. In the data phase, diverse data are collected about past patients, treatments used, and outcomes. Innovative information technologies that support semantic interoperability enable distributed learning and data sharing without additional burden on health care professionals and without the need for data to leave the hospital. In the knowledge phase, prediction models are developed for new data and treatment outcomes by applying machine learning methods to data. In the application phase, this knowledge is applied in clinical practice via novel decision support systems or via extensions of existing models such as Tumour Control Probability models. In the evaluation phase, the predictability of treatment outcomes allows the new knowledge to be evaluated by comparing predicted and actual outcomes. ConclusionPersonalised or tailored cancer therapy ensures not only that patients receive an optimal treatment, but also that the right resources are being used for the right patients. Rapid Learning approaches combined with evidence based medicine are expected to improve the predictability of outcome and radiotherapy is the ideal field to study the value of Rapid Learning. The next step will be to include patient preferences in the decision making.

Comparative Effectiveness Research, Genomics-Enabled Personalized Medicine, and Rapid Learning Health Care: A Common Bond

Despite stunning advances in our understanding of the genetics and the molecular basis for cancer, many patients with cancer are not yet receiving therapy tailored specifically to their tumor biology. The translation of these advances into clinical practice has been hindered, in part, by the lack of evidence for biomarkers supporting the personalized medicine approach. Most stakeholders agree that the translation of biomarkers into clinical care requires evidence of clinical utility. The highest level of evidence comes from randomized controlled clinical trials (RCTs). However, in many instances, there may be no RCTs that are feasible for assessing the clinical utility of potentially valuable genomic biomarkers. In the absence of RCTs, evidence generation will require well-designed cohort studies for comparative effectiveness research (CER) that link detailed clinical information to tumor biology and genomic data. CER also uses systematic reviews, evidence-quality appraisal, and health outcomes research to provide a methodologic framework for assessing biologic patient subgroups. Rapid learning health care (RLHC) is a model in which diverse data are made available, ideally in a robust and real-time fashion, potentially facilitating CER and personalized medicine. Nonetheless, to realize the full potential of personalized care using RLHC requires advances in CER and biostatistics methodology and the development of interoperable informatics systems, which has been recognized by the National Cancer Institute's program for CER and personalized medicine. The integration of CER methodology and genomics linked to RLHC should enhance, expedite, and expand the evidence generation required for fully realizing personalized cancer care.

Improving the management of dyspnea in the community using rapid learning approaches

Patients with chronic pulmonary disease often suffer from breathlessness or dyspnea. Traditional evidence generation techniques to expand upon current treatment paradigms are limited by the significant delay between study initiation and clinical implementation of findings. Rapid learning health care is a novel approach to health care delivery that relies on intelligent and continuous integration of clinical and research data sets to deliver personalized medicine using the most current evidence available. Results of important studies in the management of chronic respiratory disease are presented in brief; however, the focus of this review is on evidence supporting the implementation of a rapid learning model for symptom management. Recent findings suggest that a rapid learning system is feasible and acceptable to patients with advanced illness, helps monitor symptoms overtime, facilitates study of the impact of novel interventions, and can identify unrecognized needs and concerns. A rapid learning model improves comprehensive assessment, timeliness of intervention, and accrual of contemporaneous data to support best practice that tailors care specific to the needs of patients as their disease and lifestyle change overtime. Using the rapid learning health care model, data collected in the process of routine care can simultaneously function both as clinical information and as a resource for research on patient-centered experiences and outcomes.

Management of dyspnea within a rapid learning healthcare model

This review addresses a distressing symptom experienced by many palliative care patients, for which available interventions have been only partially effective. A new model of healthcare delivery and research - rapid learning healthcare - provides a potential framework for improving clinical care for and outcomes of dyspnea. This review places dyspnea management in palliative care within the new systems approach offered by rapid learning healthcare. Results of important studies in dyspnea are briefly presented, though the focus of this review is on evidence supporting implementation of a rapid learning model for palliative symptom management. Recent findings suggest that a rapid learning system is feasible and acceptable to patients with advanced illness, helps monitor symptoms over time, facilitates study of the impact of novel interventions, and can identify unrecognized needs and concerns. A rapid learning model could improve comprehensive assessment, timeliness of intervention, accrual of data to support best practice, and tailoring of care to patients' needs as their disease and experiences change over time. Data collected in the process of routine care in a rapid learning model can simultaneously function as clinical information and a resource for research on patient-centered experiences and outcomes.

Technology's role in healthcare reform

Technology's role in healthcare reform