Vision Transformer-based Decision Support for Neurosurgical Intervention in Acute Traumatic Brain Injury: Automated Surgical Intervention Support Tool

Purpose To develop an automated triage tool to predict neurosurgical intervention for patients with traumatic brain injury (TBI). Materials and Methods A provincial trauma registry was reviewed to retrospectively identify patients with TBI from 2005 to 2022 treated at a specialized Canadian trauma center. Model training, validation, and testing were performed using head CT scans with binary reference standard patient-level labels corresponding to whether the patient received neurosurgical intervention. Performance and accuracy of the model, the Automated Surgical Intervention Support Tool for TBI (ASIST-TBI), were also assessed using a held-out consecutive test set of all patients with TBI presenting to the center between March 2021 and September 2022. Results Head CT scans from 2806 patients with TBI (mean age, 57 years ± 22 [SD]; 1955 [70%] men) were acquired between 2005 and 2021 and used for training, validation, and testing. Consecutive scans from an additional 612 patients (mean age, 61 years ± 22; 443 [72%] men) were used to assess the performance of ASIST-TBI. There was accurate prediction of neurosurgical intervention with an area under the receiver operating characteristic curve (AUC) of 0.92 (95% CI: 0.88, 0.94), accuracy of 87% (491 of 562), sensitivity of 87% (196 of 225), and specificity of 88% (295 of 337) on the test dataset. Performance on the held-out test dataset remained robust with an AUC of 0.89 (95% CI: 0.85, 0.91), accuracy of 84% (517 of 612), sensitivity of 85% (199 of 235), and specificity of 84% (318 of 377). Conclusion A novel deep learning model was developed that could accurately predict the requirement for neurosurgical intervention using acute TBI CT scans. Keywords: CT, Brain/Brain Stem, Surgery, Trauma, Prognosis, Classification, Application Domain, Traumatic Brain Injury, Triage, Machine Learning, Decision Support Supplemental material is available for this article. © RSNA, 2024 See also commentary by Haller in this issue.

User Perceptions and Use of Decision Support Medical Apps Among Medical Students: Cross-Sectional Study

We aimed to assess medical students' use of decision-support medical apps and evaluate their perception of app use. A cross-sectional multi-center observational study was conducted among medical students with and without a medical informatics course as part of their undergraduate medical curriculum. We assessed trust, perceptions, patient impression, reliability, and comfort using an online survey. A total of 439 responses were received. There were significant differences between the two groups when indicating which apps, they trust. Students agreed that using apps enhanced knowledge (91%), saved time (88%), improved patient care (85%), and increased diagnostic accuracy (82%). Students indicated that patients would think that students didn't know what they were doing (63%) or students were fresh out of training (53%) when using apps in the presence of patients. Incorporating medical app usage as part of learning may increase trust and comfort with using medical apps in medical practice.

Advances in the Application of Traditional Chinese Medicine Using Artificial Intelligence: A Review

Traditional Chinese medicine (TCM), as one of the crystallizations of Chinese wisdom, emphasizes the balance of Yin and Yang to keep the body healthy. Under the theoretical guidance of a holistic view, the diagnostic process in TCM has characteristics of subjectivity, fuzziness, and complexity. Therefore, realizing standardization and achieving objective quantitative analysis are the bottlenecks of the development of TCM. The emergence of artificial intelligence (AI) technology has brought unprecedented challenges and opportunities to traditional medicine, which is expected to provide objective measurements and improve the clinical efficacy. However, the combination of TCM and AI is still in its infancy and currently faces many challenges. Therefore, this review provides a comprehensive discussion of the existing advances, problems, and prospects of the applications of AI technologies in TCM with the hope of promoting a better understanding of the TCM modernization and intellectualization.

Developing AI enabled sensors and decision support for military operators in the field

Wearable sensors enable down range data collection of physiological and cognitive performance of the warfighter. However, autonomous teams may find the sensor data impractical to interpret and hence influence real-time decisions without the support of subject matter experts. Decision support tools can reduce the burden of interpreting physiological data in the field and incorporate a systems perspective where noisy field data can contain useful additional signals. We present a methodology of how artificial intelligence can be used for modeling human performance with decision-making to achieve actionable decision support. We provide a framework for systems design and advancing from the laboratory to real world environments. The result is a validated measure of down-range human performance with a low burden of operation.

Climate-related risk modeling of Banana Xanthomonas Wilt (BXW) disease incidence within cropland area of Rwanda

Banana Xanthomonas Wilt (BXW) is the major threat to banana in Rwanda, causing up to 100% yield loss. There are no biological or chemical control measures, and little is known about potential direction and magnitude of its spread, so cultural control efforts are reactive rather than proactive. In this study, we assessed BXW risk under current and projected climate, to guide early warning and control by applying maximum entropy (Maxent) model on 1,022 georeferenced BXW datapoints and 20 environmental variables. We evaluated the variables' significance and mapped potential risk under current and future climates to assess spatial dynamics of the disease distribution. BXW occurrence was reliably predicted (mean validation AUC ranging from 0.79 - 0.85). Precipitation of the coldest quarter, average maximum monthly temperature, annual precipitation, and elevation were strongest predictors, explaining 21.1%, 13%, 12.6%, and 9.4% of the observed incidence variability, respectively, while mean temperature of the coldest quarter has the highest gain in isolation. Further, the most susceptible regions (in western, northern, and southern Rwanda) are characterized by elevation (1,350m - 2,000m), annual precipitation (900mm - 1700mm), and average temperature (14oC - 20oC), among other variables, suggesting that a consistent, rainy, and warm climate is more favorable for BXW spread. Under future climate, the risk is predicted to increase and spread to other regions. We conclude that climate change will likely exacerbate BXW-related losses of banana land area and yield, under temperature and precipitation influence. Findings support evidence-based targeting of extension delivery to farmers and national early warning for timely action.

Family planning decisional needs assessment for recessive hereditary disorders: Insights from carrier couples and professionals

OBJECTIVE

Carrier couples of recessive diseases face an important decision-making process regarding their familial planning that can become a significant source of discomfort and potential regrets. To date, no study has described the decisional needs of carrier couples of Leigh syndrome French-Canadian type, hereditary tyrosinemia type 1, hereditary motor and sensory neuropathy with or without agenesis of the corpus callosum and autosomal recessive spastic ataxia of Charlevoix-Saguenay undergoing family planning decision-making process. Our study aimed to describe the decisional needs of carrier couples according to the Ottawa Decision Support Framework.

METHODS

A qualitative descriptive study was conducted. Qualitative individual and joint couple interviews were performed among 39 carrier individuals and 11 health and social care professionals.

RESULTS

Carrier couples' decision-making process is complex, and their decisional needs include, among others, sufficient knowledge about the disease and the reproductive options, personal values and available support. Increased decisional conflict was observed among carriers of diseases associated with low morbidity and mortality.

CONCLUSION

Family planning decision-making is an evolutive deliberative process that requires appropriate shared decision-making to insure decisional comfort.

PRACTICE IMPLICATIONS

Carrier couples family planning decision-making process needs to be better supported.

Comparison of MetaMap, cTAKES, SIFR, and ECMT to Annotate Breast Cancer Patient Summaries

Most clinical texts including breast cancer patient summaries (BCPSs) are elaborated as narrative documents difficult to process by decision support systems. Annotators have been developed to extract the relevant content of such documents, e.g., MetaMap and cTAKES, that work with the English language and perform concept mapping using UMLS, SIFR and ECMT, that work for the French language and provide concepts using various terminologies. We compared the four annotators on a sample of 25 French BCPSs, pre-processed to manage acronyms and translated in English. We observed that MetaMap extracted the largest number of UMLS concepts (15,458), followed by SIFR (3,784), ECMT (1,962), and cTAKES (1,769). Each annotator extracted specific valuable information, not proposed by the other annotators. Considered as complementary, all annotators should be used in sequence to optimize the results.

Exploiting the Capabilities of Bayesian Networks for Engineering Risk Assessment: Causal Reasoning through Interventions

In the last decade, Bayesian networks (BNs) have been widely used in engineering risk assessment due to the benefits that they provide over other methods. Among these, the most significant is the ability to model systems, causal factors, and their dependencies in a probabilistic manner. This capability has enabled the community to do causal reasoning through associations, which answers questions such as: "How does new evidence x'$x^{\prime }$ about the occurrence of event X$X$ change my belief about the occurrence of event Y$Y$ ?" Associative reasoning has helped risk analysts to identify relevant risk-contributing factors and perform scenario analysis by evidence propagation. However, engineering risk assessment has yet to explore other features of BNs, such as the ability to reason through interventions, which enables the BN model to support answering questions of the form "How does doing X=x'$X=x^{\prime }$ change my belief about the occurrence of event Y$Y$ ?" In this article, we propose to expand the scope of use of BN models in engineering risk assessment to support intervention reasoning. This will provide more robust risk-informed decision support by enabling the modeling of policies and actions before being implemented. To do this, we provide the formal mathematical background and tools to model interventions in BNs and propose a framework that enables its use in engineering risk assessment. This is demonstrated in an illustrative case study on third-party damage of natural gas pipelines, showing how BNs can be used to inform decision-makers about the effect that new actions/policies can have on a system.

Are Semantic Annotators Able to Extract Relevant Complexity-Related Concepts from Clinical Notes?

Clinical decision support systems (CDSSs) implementing cancer clinical practice guidelines (CPGs) have the potential to improve the compliance of decisions made by multidisciplinary tumor boards (MTB) with CPGs. However, guideline-based CDSSs do not cover complex cases and need time for discussion. We propose to learn how to predict complex cancer cases prior to MTBs from breast cancer patient summaries (BCPSs) resuming clinical notes. BCPSs being unstructured natural language textual documents, we implemented four semantic annotators (ECMT, SIFR, cTAKES, and MetaMap) to assess whether complexity-related concepts could be extracted from clinical notes. On a sample of 24 BCPSs covering 35 complexity reasons, ECMT and MetaMap were the most efficient systems with a performance rate of 60% (21/35) and 49% (17/35), respectively. When using the four annotators in sequence, 69% of complexity reasons were extracted (24/35 reasons).

Design of a novel multifunction decision support/alerting system for in-patient acute care, ICU and floor (AlertWatch AC)

Background

Multifunction surveillance alerting systems have been found to be beneficial for the operating room and labor and delivery. This paper describes a similar system developed for in-hospital acute care environments, AlertWatch Acute Care (AWAC).

Results

A decision support surveillance system has been developed which extracts comprehensive electronic health record (EHR) data including live data from physiologic monitors and ventilators and incorporates them into an integrated organ icon-based patient display. Live data retrieved from the hospitals network are processed by presenting scrolling median values to reduce artifacts. A total of 48 possible alerts are generated covering a broad range of critical patient care concerns. Notification is achieved by paging or texting the appropriated member of the critical care team. Alerts range from simple out of range values to more complex programing of impending Ventilator Associated Events, SOFA, qSOFA, SIRS scores and process of care reminders for the management of glucose and sepsis. As with similar systems developed for the operating room and labor and delivery, there are green, yellow, and red configurable ranges for all parameters. A census view allows surveillance of an entire unit with flashing or text to voice alerting and enables detailed information by windowing into an individual patient view including live physiologic waveforms. The system runs via web interface on desktop as well as mobile devices, with iOS native app available, for ease of communication from any location. The goal is to improve safety and adherence to standard management protocols.

Conclusions

AWAC is designed to provide a high level surveillance view for multi-bed hospital units with varying acuity from standard floor patients to complex ICU care. Alerts are generated by algorithms running in the background and automatically notify the selected member of the patients care team. Its value has been demonstrated for low acuity patients, further study is required to determine its effectiveness in high acuity patients.

Beyond Two Cultures: Cultural Infrastructure for Data-driven Decision Support

In the twenty years since Dr. Leo Breiman's incendiary paper Statistical Modeling: The Two Cultures was first published, algorithmic modeling techniques have gone from controversial to commonplace in the statistical community. While the widespread adoption of these methods as part of the contemporary statistician's toolkit is a testament to Dr. Breiman's vision, the number of high-profile failures of algorithmic models suggests that Dr. Breiman's final remark that "the emphasis needs to be on the problem and the data" has been less widely heeded. In the spirit of Dr. Breiman, we detail an emerging research community in statistics - data-driven decision support. We assert that to realize the full potential of decision support, broadly and in the context of precision health, will require a culture of social awareness and accountability, in addition to ongoing attention towards complex technical challenges.

Decision-Making Experiences Of Consumers Choosing Individual-Market Health Insurance Plans

The health insurance Marketplaces established by the Affordable Care Act include features designed to simplify the process of choosing a health plan in the individual, or nongroup, insurance market. While most individual health insurance enrollees purchase plans through the federal and state-based Marketplaces, millions also purchase plans directly from an insurance carrier (off Marketplace). This study was a descriptive comparison of the decision-making processes and shopping experiences of consumers in two states who purchased a health insurance plan from the same large insurer in 2017, either through the federal Marketplaces or off Marketplace. In a survey, those who selected plans through the Marketplaces reported less difficulty finding the best or most affordable plan than did those enrolling off Marketplace. Respondents in families with chronic health conditions who enrolled through the Marketplaces reported better overall experiences than those who enrolled off Marketplace. Respondents with low health insurance literacy reported poor experiences in enrolling both through the Marketplaces and off Marketplace. Access to consumer assistance in the individual health insurance market should target off-Marketplace populations as well as all populations with low health insurance literacy.

Protocol for the Effectiveness of an Anesthesiology Control Tower System in Improving Perioperative Quality Metrics and Clinical Outcomes: the TECTONICS randomized, pragmatic trial

Introduction: Perioperative morbidity is a public health priority, and surgical volume is increasing rapidly. With advances in technology, there is an opportunity to research the utility of a telemedicine-based control center for anesthesia clinicians that assess risk, diagnoses negative patient trajectories, and implements evidence-based practices. Objectives: The primary objective of this trial is to determine whether an anesthesiology control tower (ACT) prevents clinically relevant adverse postoperative outcomes including 30-day mortality, delirium, respiratory failure, and acute kidney injury. Secondary objectives are to determine whether the ACT improves perioperative quality of care metrics including management of temperature, mean arterial pressure, mean airway pressure with mechanical ventilation, blood glucose, anesthetic concentration, antibiotic redosing, and efficient fresh gas flow. Methods and analysis: We are conducting a single center, randomized, controlled, phase 3 pragmatic clinical trial. A total of 58 operating rooms are randomized daily to receive support from the ACT or not. All adults (eighteen years and older) undergoing surgical procedures in these operating rooms are included and followed until 30 days after their surgery. Clinicians in operating rooms randomized to ACT support receive decision support from clinicians in the ACT. In operating rooms randomized to no intervention, the current standard of anesthesia care is delivered. The intention-to-treat principle will be followed for all analyses. Differences between groups will be presented with 99% confidence intervals; p-values <0.005 will be reported as providing compelling evidence, and p-values between 0.05 and 0.005 will be reported as providing suggestive evidence. Registration: TECTONICS is registered on ClinicalTrials.gov, NCT03923699; registered on 23 April 2019.

The effect of imputing missing clinical attribute values on training lung cancer survival prediction model performance

According to the estimations of the World Health Organization and the International Agency for Research in Cancer, lung cancer is the most common cause of death from cancer worldwide. The last few years have witnessed a rise in the attention given to the use of clinical decision support systems in medicine generally and in cancer in particular. These can predict patients' likelihood of survival based on analysis of and learning from previously treated patients. The datasets that are mined for developing clinical decision support functionality are often incomplete, which adversely impacts the quality of the models developed and the decision support offered. Imputing missing data using a statistical analysis approach is a common method to addressing the missing data problem. This work investigates the effect of imputation methods for missing data in preparing a training dataset for a Non-Small Cell Lung Cancer survival prediction model using several machine learning algorithms. The investigation includes an assessment of the effect of imputation algorithm error on performance prediction and also a comparison between using a smaller complete real dataset or a larger dataset with imputed data. Our results show that even when the proportion of records with some missing data is very high (> 80%) imputation can lead to prediction models with an AUC (0.68-0.72) comparable to those trained with complete data records.

Impact of home care versus alternative locations of care on elder health outcomes: an overview of systematic reviews

Background

Many elders struggle with the decision to remain at home or to move to an alternative location of care. A person’s location of care can influence health and wellbeing. Healthcare organizations and policy makers are increasingly challenged to better support elders’ dwelling and health care needs. A summary of the evidence that examines home care compared to other care locations can inform decision making. We surveyed and summarized the evidence evaluating the impact of home care versus alternative locations of care on elder health outcomes.

Methods

We conducted an overview of systematic reviews. Data sources included MEDLINE, the Cochrane Library, EMBASE, and CINAHL. Eligible reviews included adults 65+ years, elder home care, alternative care locations, and elder health outcomes. Two independent reviewers screened citations. We extracted data and appraised review quality using the Assessing the Methodological Quality of Systematic Reviews (AMSTAR) checklist. Results were synthesized narratively.

Results

The search yielded 2575 citations, of which 19 systematic reviews were eligible. Three hundred and forty studies with 271,660 participants were synthesized across the systematic reviews. The categories of comparisons included: home with support versus independent living at home (n = 11 reviews), home care versus institutional care (n = 3 reviews), and rehabilitation at home versus conventional rehabilitation services (n = 7 reviews). Two reviews had data relevant to two categories. Most reviews favoured home with support to independent living at home. Findings comparing home care to institutional care were mixed. Most reviews found no differences in health outcomes between rehabilitation at home versus conventional rehabilitation services. Systematic review quality was moderate, with a median AMSTAR score of 6 (range 4 - 10 out of 11).

Conclusions

The evidence on the impact of home care compared to alternative care locations on elder health outcomes is heterogeneous. Our findings support positive health impacts of home support interventions for community dwelling elders compared to independent living at home. There is insufficient evidence to determine the impact of alternative care locations on elders’ health. Additional research targeting housing and care options for the elderly is needed.

Electronic supplementary material

The online version of this article (doi:10.1186/s12877-016-0395-y) contains supplementary material, which is available to authorized users.

A Decision Support Tool for Sustainable Land Use, Transportation, Buildings/Infrastructure, and Materials Management

One issue for community groups, local and regional planners, and politicians, is that they require relevant information to develop programs and initiatives for incorporating sustainability principles into their physical infrastructure, operations, and decision-making processes. This research project addressed the issue through two research questions. The first research question that this project was designed to address, which was greatly influenced by the vast number of references in the sustainability literature, was an ontological one, " what are the major categories that sustainability decisions can be grouped under, and how might those categories be related? ". The second research question that this project answered was, " how can information contained in the sustainability literature be made accessible to users in a convenient format? ". The Multi-Sector Sustainability Browser (MSSB) is a decision support tool (DST) designed to synthesize and summarize research in four sustainability decision domains, Land Use, Buildings and Infrastructure, Transportation, and Materials Management in a manner that provides easy and rapid access to information for use in planning and decision making. Weblinks are provided to reference documents and resources from the four sustainability decision domains, allowing users to download relevant documents and extract information in support of sustainability decisions and related program initiatives.

A Digital Framework to Support Providers and Patients in Diabetes Related Behavior Modification

We present Diabetes Web-Centric Information and Support Environment (D-WISE) that features: (a) Decision support tool to assist family physicians to administer Behavior Modification (BM) strategies to patients; and (b) Patient BM application that offers BM strategies and motivational interventions to engage patients. We take a knowledge management approach, using semantic web technologies, to model the social cognition theory constructs, Canadian diabetes guidelines and BM protocols used locally, in terms of a BM ontology that drives the BM decision support to physicians and BM strategy adherence monitoring and messaging to patients. We present the qualitative analysis of D-WISE usability by both physicians and patients.

Cross-Cultural Adaptation of a Decision Support System for AT Selection

Older adults may benefit from decision support systems for the selection of assistive technologies. We present a case study cross-cultural adaptation in Canada of a system developed in the United Kingdom. We describe a conceptual framework for examining challenges with cultural and literacy issues and ways to address them.

Decision-Tree Model for Support of Health Policy Choices Based on Pneumococcal Conjugate Vaccine (PCV) Program Outcomes

Pneumococcal Conjugate Vaccine (PCV) has the potential to save lives in low-income countries. We have developed a computational model and web-based decision support software for comparing cost-benefit tradeoffs from alternative PCV program designs, considering their direct and indirect effects on early childhood populations in resource-poor settings. This supports policy-makers in estimating potential health outcomes and cost-effectiveness of different vaccination program strategies for a wide range of population coverage and vaccine effectiveness assumptions.

Evaluation of a Decision Support System for Laboratory Service Patients

In Russia, many patients address laboratory services directly without a doctor's referral. This causes the problem of interpretation of laboratory test results by the patients who don't have a proper medical background. We have a situation when there is no healthcare professional between a patient and the test results. This problem can be solved if a laboratory service provides not only the results of the tests but also their interpretations. The system was implemented in the Helix laboratory service in Saint-Petersburg, Russia. At the moment it generates about 3500 reports a day. This study presents the evaluation of the system, which has shown a good performance both on correctness of generated reports and user acceptance.

Data-driven decision-making tools to improve public resource allocation for care and prevention of HIV/AIDS

Public health agencies face difficult decisions when allocating scarce resources to control the spread of HIV/AIDS. Decisions are often made with few local empirical data. We demonstrated the use of the robust decision making approach in Los Angeles County, an approach that is data driven and allows decision makers to compare the performance of various intervention strategies across thousands of simulated future scenarios. We found that the prevailing strategy of emphasizing behavioral risk reduction interventions was unlikely to achieve the policy goals of the national HIV/AIDS strategy. Of the alternative strategies we examined, those that invested most heavily in interventions to initiate antiretroviral treatment and support treatment adherence were the most likely to achieve policy objectives. By employing similar methods, other public health agencies can identify robust strategies and invest in interventions more likely to achieve HIV/AIDS policy goals.

Evaluation of potential novel variations and their interactions related to bipolar disorders: analysis of genome-wide association study data

BACKGROUND

Multifactor dimensionality reduction (MDR) is a nonparametric approach that can be used to detect relevant interactions between single-nucleotide polymorphisms (SNPs). The aim of this study was to build the best genomic model based on SNP associations and to identify candidate polymorphisms that are the underlying molecular basis of the bipolar disorders.

METHODS

This study was performed on Whole-Genome Association Study of Bipolar Disorder (dbGaP [database of Genotypes and Phenotypes] study accession number: phs000017.v3.p1) data. After preprocessing of the genotyping data, three classification-based data mining methods (ie, random forest, naïve Bayes, and k-nearest neighbor) were performed. Additionally, as a nonparametric, model-free approach, the MDR method was used to evaluate the SNP profiles. The validity of these methods was evaluated using true classification rate, recall (sensitivity), precision (positive predictive value), and F-measure.

RESULTS

Random forests, naïve Bayes, and k-nearest neighbors identified 16, 13, and ten candidate SNPs, respectively. Surprisingly, the top six SNPs were reported by all three methods. Random forests and k-nearest neighbors were more successful than naïve Bayes, with recall values >0.95. On the other hand, MDR generated a model with comparable predictive performance based on five SNPs. Although different SNP profiles were identified in MDR compared to the classification-based models, all models mapped SNPs to the DOCK10 gene.

CONCLUSION

Three classification-based data mining approaches, random forests, naïve Bayes, and k-nearest neighbors, have prioritized similar SNP profiles as predictors of bipolar disorders, in contrast to MDR, which has found different SNPs through analysis of two-way and three-way interactions. The reduced number of associated SNPs discovered by MDR, without loss in the classification performance, would facilitate validation studies and decision support models, and would reduce the cost to develop predictive and diagnostic tests. Nevertheless, we need to emphasize that translation of genomic models to the clinical setting requires models with higher classification performance.

Focused Decision Support: a Data Mining Tool to Query the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Dataset and Guide Screening Management for the Individual Patient

The Prostate, Lung, Colorectal, and Ovarian Cancer (PLCO) Screening Trial enrolled ~155,000 participants to determine whether certain screening exams reduced mortality from prostate, lung, colorectal, and ovarian cancer. Repurposing the data provides an unparalleled resource for matching patients with the outcomes of demographically or diagnostically comparable patients. A web-based application was developed to query this subset of patient information against a given patient's demographics and risk factors. Analysis of the matched data yields outcome information which can then be used to guide management decisions and imaging software. Prognostic information is also estimated via the proportion of matched patients that progress to cancer. The US Preventative Services Task Force provides screening recommendations for cancers of the breast, colorectal tract, and lungs. There is wide variability in adherence of clinicians to these guidelines and others published by the Fleischner Society and various cancer organizations. Data mining the PLCO dataset for clinical decision support can optimize the use of limited healthcare resources, focusing screening on patients for whom the benefit to risk ratio is the greatest and most efficacious. A data driven, personalized approach to cancer screening maximizes the economic and clinical efficacy and enables early identification of patients in which the course of disease can be improved. Our dynamic decision support system utilizes a subset of the PLCO dataset as a reference model to determine imaging and testing appropriateness while offering prognostic information for various cancers.

Using electronic health record data to develop inpatient mortality predictive model: Acute Laboratory Risk of Mortality Score (ALaRMS)

Objective

Using numeric laboratory data and administrative data from hospital electronic health record (EHR) systems, to develop an inpatient mortality predictive model.

Methods

Using EHR data of 1 428 824 adult discharges from 70 hospitals in 2006–2007, we developed the Acute Laboratory Risk of Mortality Score (ALaRMS) using age, gender, and initial laboratory values on admission as candidate variables. We then added administrative variables using the Agency for Healthcare Research and Quality (AHRQ)'s clinical classification software (CCS) and comorbidity software (CS) as disease classification tools. We validated the model using 770 523 discharges in 2008.

Results

Mortality predictors with ORs >2.00 included age, deranged albumin, arterial pH, bands, blood urea nitrogen, oxygen partial pressure, platelets, pro-brain natriuretic peptide, troponin I, and white blood cell counts. The ALaRMS model c-statistic was 0.87. Adding the CCS and CS variables increased the c-statistic to 0.91. The relative contributions were 69% (ALaRMS), 25% (CCS), and 6% (CS). Furthermore, the integrated discrimination improvement statistic demonstrated a 127% (95% CI 122% to 133%) overall improvement when ALaRMS was added to CCS and CS variables. In contrast, only a 22% (CI 19% to 25%) improvement was seen when CCS and CS variables were added to ALaRMS.

Conclusions

EHR data can generate clinically plausible mortality predictive models with excellent discrimination. ALaRMS uses automated laboratory data widely available on admission, providing opportunities to aid real-time decision support. Models that incorporate laboratory and AHRQ's CCS and CS variables have utility for risk adjustment in retrospective outcome studies.

Medical decision support using machine learning for early detection of late-onset neonatal sepsis

OBJECTIVE

The objective was to develop non-invasive predictive models for late-onset neonatal sepsis from off-the-shelf medical data and electronic medical records (EMR).

DESIGN

The data used in this study are from 299 infants admitted to the neonatal intensive care unit in the Monroe Carell Jr. Children's Hospital at Vanderbilt and evaluated for late-onset sepsis. Gold standard diagnostic labels (sepsis negative, culture positive sepsis, culture negative/clinical sepsis) were assigned based on all the laboratory, clinical and microbiology data available in EMR. Only data that were available up to 12 h after phlebotomy for blood culture testing were used to build predictive models using machine learning (ML) algorithms.

MEASUREMENT

We compared sensitivity, specificity, positive predictive value and negative predictive value of sepsis treatment of physicians with the predictions of models generated by ML algorithms.

RESULTS

The treatment sensitivity of all the nine ML algorithms and specificity of eight out of the nine ML algorithms tested exceeded that of the physician when culture-negative sepsis was included. When culture-negative sepsis was excluded both sensitivity and specificity exceeded that of the physician for all the ML algorithms. The top three predictive variables were the hematocrit or packed cell volume, chorioamnionitis and respiratory rate.

CONCLUSIONS

Predictive models developed from off-the-shelf and EMR data using ML algorithms exceeded the treatment sensitivity and treatment specificity of clinicians. A prospective study is warranted to assess the clinical utility of the ML algorithms in improving the accuracy of antibiotic use in the management of neonatal sepsis.

Telerehabilitation technologies: accessibility and usability

In the fields of telehealth and telemedicine, phone and/or video technologies are key to the successful provision of services such as remote monitoring and visits. How do these technologies affect service accessibility, effectiveness, quality, and usefulness when applied to rehabilitation services in the field of telerehabilitation? To answer this question, we provide a overview of the complex network of available technologies and discuss how they link to rehabilitation applications, services, and practices as well as to the telerehabilitation end-user. This white paper will first present the numerous professional considerations that shape the use of technology in telerehabilitation service and set it somewhat apart from telemedicine. It will then provide an overview of concepts essential to usability analysis; present a summary of various telerehabilitation technologies and their strengths and limitations, and consider how the technologies interface with end users’ clinical needs for service accessibility, effectiveness, quality, and usefulness. The paper will highlight a conceptual framework (including task analyses and usability issues) that underlies a functional match between telerehabilitation technologies, clinical applications, and end-user capabilities for telerehabilitation purposes. Finally, we will discuss pragmatic issues related to user integration of telerehabilitation technology versus traditional face-to-face approaches.