Fernanda Polubriaginof

The feasibility of using natural language processing to extract clinical information from breast pathology reports.

Objective: The opportunity to integrate clinical decision support systems into clinical practice is limited due to the lack of structured, machine readable data in the current format of the electronic health record. Natural language processing has been designed to convert free text into machine readable data. The aim of the current study was to ascertain the feasibility of using natural language processing to extract clinical information from >76,000 breast pathology reports. Approach and Procedure: Breast pathology reports from three institutions were analyzed using natural language processing software (Clearforest, Waltham, MA) to extract information on a variety of pathologic diagnoses of interest. Data tables were created from the extracted information according to date of surgery, side of surgery, and medical record number. The variety of ways in which each diagnosis could be represented was recorded, as a means of demonstrating the complexity of machine interpretation of free text. Results: There was widespread variation in how pathologists reported common pathologic diagnoses. We report, for example, 124 ways of saying invasive ductal carcinoma and 95 ways of saying invasive lobular carcinoma. There were >4000 ways of saying invasive ductal carcinoma was not present. Natural language processor sensitivity and specificity were 99.1% and 96.5% when compared to expert human coders. Conclusion: We have demonstrated how a large body of free text medical information such as seen in breast pathology reports, can be converted to a machine readable format using natural language processing, and described the inherent complexities of the task.

Using machine learning to parse breast pathology reports

Purpose Extracting information from electronic medical record is a time-consuming and expensive process when done manually. Rule-based and machine learning techniques are two approaches to solving this problem. In this study, we trained a machine learning model on pathology reports to extract pertinent tumor characteristics, which enabled us to create a large database of attribute searchable pathology reports. This database can be used to identify cohorts of patients with characteristics of interest.Methods We collected a total of 91,505 breast pathology reports from three Partners hospitals: Massachusetts General Hospital, Brigham and Women’s Hospital, and Newton-Wellesley Hospital, covering the period from 1978 to 2016. We trained our system with annotations from two datasets, consisting of 6295 and 10,841 manually annotated reports. The system extracts 20 separate categories of information, including atypia types and various tumor characteristics such as receptors. We also report a learning curve analysis to show how much annotation our model needs to perform reasonably.Results The model accuracy was tested on 500 reports that did not overlap with the training set. The model achieved accuracy of 90% for correctly parsing all carcinoma and atypia categories for a given patient. The average accuracy for individual categories was 97%. Using this classifier, we created a database of 91,505 parsed pathology reports.ConclusionsOur learning curve analysis shows that the model can achieve reasonable results even when trained on a few annotations. We developed a user-friendly interface to the database that allows physicians to easily identify patients with target characteristics and export the matching cohort. This model has the potential to reduce the effort required for analyzing large amounts of data from medical records, and to minimize the cost and time required to glean scientific insight from these data.

Breast Cancer Risk and Follow-up Recommendations for Young Women Diagnosed with Atypical Hyperplasia and Lobular Carcinoma In Situ (LCIS).

BackgroundThe risk of breast cancer in young women diagnosed with atypical hyperplasia and (LCIS) is not well defined. The objectives were to evaluate outcomes and to help determine guidelines for follow-up in this population.MethodsA retrospective review of women under age 35 diagnosed with ADH, ALH, LCIS, and severe ADH from 1987 to 2010 was performed. Patient characteristics, pathology and follow-up were determined from chart review.ResultsWe identified 58 young women with atypical breast lesions. Median age at diagnosis was 31 years (range 19–34). 34 patients had ADH, 11 had ALH, 8 had LCIS, and 5 had severe ADH.7 (12%) patients developed breast cancer. The median follow-up was 86 months (range 1–298). Median time to cancer diagnosis was 90 months (range 37–231). 4 cancers were on the same side, 3 were contralateral. 4 were IDC, 1 was ILC, and 2 were DCIS.Cancer was detected by screening mammogram in 4 patients, 2 by clinical exam, and 1 unknown. In the entire cohort, 26 (45%) patients had screening mammograms as part of their follow up, 12 patients had only clinical follow up, and 20 had no additional follow up. 13 patients required subsequent biopsies.ConclusionYoung women with atypical breast lesions are at a markedly increased risk for developing breast cancer and should be followed closely. Based on our findings, we recommend close clinical follow-up, MRI starting at age 25 through age 29, and screening mammograms for those over 30 in this high-risk group of patients.

Disease Heritability Inferred from Familial Relationships Reported in Medical Records

Heritability is essential for understanding the biological causes of disease but requires laborious patient recruitment and phenotype ascertainment. Electronic health records (EHRs) passively capture a wide range of clinically relevant data and provide a resource for studying the heritability of traits that are not typically accessible. EHRs contain next-of-kin information collected via patient emergency contact forms, but until now, these data have gone unused in research. We mined emergency contact data at three academic medical centers and identified 7.4 million familial relationships while maintaining patient privacy. Identified relationships were consistent with genetically derived relatedness. We used EHR data to compute heritability estimates for 500 disease phenotypes. Overall, estimates were consistent with the literature and between sites. Inconsistencies were indicative of limitations and opportunities unique to EHR research. These analyses provide a validation of the use of EHRs for genetics and disease research.

Procedure prediction from symbolic Electronic Health Records via time intervals analytics

Prediction of medical events, such as clinical procedures, is essential for preventing disease, understanding disease mechanism, and increasing patient quality of care. Although longitudinal clinical data from Electronic Health Records provides opportunities to develop predictive models, the use of these data faces significant challenges. Primarily, while the data are longitudinal and represent thousands of conceptual events having duration, they are also sparse, complicating the application of traditional analysis approaches. Furthermore, the framework presented here takes advantage of the events duration and gaps. International standards for electronic healthcare data represent data elements, such as procedures, conditions, and drug exposures, using eras, or time intervals. Such eras contain both an event and a duration and enable the application of time intervals mining - a relatively new subfield of data mining. In this study, we present Maitreya, a framework for time intervals analytics in longitudinal clinical data. Maitreya discovers frequent time intervals related patterns (TIRPs), which we use as prognostic markers for modelling clinical events. We introduce three novel TIRP metrics that are normalized versions of the horizontal-support, that represents the number of TIRP instances per patient. We evaluate Maitreya on 28 frequent and clinically important procedures, using the three novel TIRP representation metrics in comparison to no temporal representation and previous TIRPs metrics. We also evaluate the epsilon value that makes Allens relations more flexible with several settings of 30, 60, 90 and 180days in comparison to the default zero. For twenty-two of these procedures, the use of temporal patterns as predictors was superior to non-temporal features, and the use of the vertically normalized horizontal support metric to represent TIRPs as features was most effective. The use of the epsilon value with thirty days was slightly better than the zero.

International perspectives on sharing clinical data with patients

OBJECTIVE Engaging patients in their care has become a topic of increasing importance, and enabling patients to have access to their clinical data is a key aspect of such engagement. We investigated, on an international scale, the current state of approaches for providing patients with access to their own clinical information. METHODS Individuals from 28 countries were invited to participate in a cross-sectional semi-structured interview. Interview questions focused on social and cultural influences that affected patient engagement activities, government support for current and planned initiatives, data ownership models, and technical issues. RESULTS Interviews were conducted with individuals from 16 countries representing six continents. Respondents reported substantive initiatives for providing information to patients in the majority of countries interviewed. These initiatives were diverse in nature and stage of implementation. DISCUSSION Enabling patient access to data is occurring on an international scale. There is considerable variability in the level of maturity, the degree of government involvement, the technical infrastructure, and the plans for future development across the world. As informaticians, we are still in the early stages of deploying patient engagement technologies and have yet to identify optimal strategies in this arena. CONCLUSION Efforts to improve patient access to data are active on a global-scale. There are many open questions about best practices and much can be learned by adopting an international perspective to guide future implementation efforts.

Engaging hospital patients in the medication reconciliation process using tablet computers

Abstract Objective Unintentional medication discrepancies contribute to preventable adverse drug events in patients. Patient engagement in medication safety beyond verbal participation in medication reconciliation is limited. We conducted a pilot study to determine whether patients’ use of an electronic home medication review tool could improve medication safety during hospitalization. Materials and Methods Patients were randomized to use a toolbefore orafter hospital admission medication reconciliation to review and modify their home medication list. We assessed the quantity, potential severity, and potential harm of patients’ and clinicians’ medication changes. We also surveyed clinicians to assess the tool’s usefulness. Results Of 76 patients approached, 65 (86%) participated. Forty-eight (74%) made changes to their home medication list [before: 29 (81%),after: 19 (66%),p = .170].Before group participants identified 57 changes that clinicians subsequently missed on admission medication reconciliation. Thirty-nine (74%) had a significant or greater potential severity, and 19 (36%) had a greater than 50-50 chance of harm.After group patients identified 68 additional changes to their reconciled medication lists. Fifty-one (75%) had a significant or greater potential severity, and 33 (49%) had a greater than 50-50 chance of harm. Clinicians reported believing that the tool would save time, and patients would supply useful information. Discussion The results demonstrate a high willingness of patients to engage in medication reconciliation, and show that patients were able to identify important medication discrepancies and often changes that clinicians missed. Conclusion Engaging patients in admission medication reconciliation using an electronic home medication review tool may improve medication safety during hospitalization.

Development of A Machine Learning Algorithm to Classify Drugs Of Unknown Fetal Effect

Many drugs commonly prescribed during pregnancy lack a fetal safety recommendation – called FDA ‘category C’ drugs. This study aims to classify these drugs into harmful and safe categories using knowledge gained from chemoinformatics (i.e., pharmacological similarity with drugs of known fetal effect) and empirical data (i.e., derived from Electronic Health Records). Our fetal loss cohort contains 14,922 affected and 33,043 unaffected pregnancies and our congenital anomalies cohort contains 5,658 affected and 31,240 unaffected infants. We trained a random forest to classify drugs of unknown pregnancy class into harmful or safe categories, focusing on two distinct outcomes: fetal loss and congenital anomalies. Our models achieved an out-of-bag accuracy of 91% for fetal loss and 87% for congenital anomalies outperforming null models. Fifty-seven ‘category C’ medications were classified as harmful for fetal loss and eleven for congenital anomalies. This includes medications with documented harmful effects, including naproxen, ibuprofen and rubella live vaccine. We also identified several novel drugs, e.g., haloperidol, that increased the risk of fetal loss. Our approach provides important information on the harmfulness of ‘category C’ drugs. This is needed, as no FDA recommendation exists for these drugs’ fetal safety.

Machine learning: novel bioinformatics approaches for combating antimicrobial resistance

Purpose of review Antimicrobial resistance (AMR) is a threat to global health and new approaches to combating AMR are needed. Use of machine learning in addressing AMR is in its infancy but has made promising steps. We reviewed the current literature on the use of machine learning for studying bacterial AMR. Recent findings The advent of large-scale data sets provided by next-generation sequencing and electronic health records make applying machine learning to the study and treatment of AMR possible. To date, it has been used for antimicrobial susceptibility genotype/phenotype prediction, development of AMR clinical decision rules, novel antimicrobial agent discovery and antimicrobial therapy optimization. Summary Application of machine learning to studying AMR is feasible but remains limited. Implementation of machine learning in clinical settings faces barriers to uptake with concerns regarding model interpretability and data quality. Future applications of machine learning to AMR are likely to be laboratory-based, such as antimicrobial susceptibility phenotype prediction.