Search

BACKGROUND Limited data accuracy is often cited as a reason for caution in the integration of physiological data obtained from consumer-oriented wearable devices in care management pathways. The effect of decreasing accuracy on predictive models generated from these data has not been previously investigated. OBJECTIVE The aim of this study is to simulate the effect of data degradation on the reliability of prediction models generated from those data and thus determine the extent to which lower device accuracy might or might not limit their use in clinical settings. METHODS Using the Multilevel Monitoring of Activity and Sleep in Healthy People data set, which includes continuous free-living step count and heart rate data from 21 healthy volunteers, we trained a random forest model to predict cardiac competence. Model performance in 75 perturbed data sets with increasing missingness, noisiness, bias, and a combination of all 3 perturbations was compared to model performance for the unperturbed data set. RESULTS The unperturbed data set achieved a mean root mean square error (RMSE) of 0.079 (SD 0.001) in predicting cardiac competence index. For all types of perturbations, RMSE remained stable up to 20%-30% perturbation. Above this level, RMSE started increasing and reached the point at which the model was no longer predictive at 80% for noise, 50% for missingness, and 35% for the combination of all perturbations. Introducing systematic bias in the underlying data had no effect on RMSE. CONCLUSIONS In this proof-of-concept study, the performance of predictive models for cardiac competence generated from continuously acquired physiological data was relatively stable with declining quality of the source data. As such, lower accuracy of consumer-oriented wearable devices might not be an absolute contraindication for their use in clinical prediction models.

Men with locally advanced prostate cancer who undergo radical prostatectomy (RP) often develop recurrence and require postoperative radiotherapy. We aimed to determine the safety of neoadjuvant stereotactic body radiotherapy (SBRT) before RP in this population.A single-institution phase 1 trial (NCT02946008) of men with high-risk or node-positive prostate cancer were enrolled between March and October 2017. The primary endpoint was to determine the maximum tolerated dose of SBRT based on a composite 30-day post-RP toxicity goal of ≤28% of patients experiencing a dose-limiting toxicity (DLT). Secondary outcomes included toxicity, efficacy, and multiple quality of life (QoL) inventories. SBRT (30-35 Gy/5 fractions) was delivered to the prostate and seminal vesicles, and 25 Gy/5 fractions to the pelvic lymph nodes. RP was performed for a median of 6 weeks post-SBRT. Hormone therapy was not allowed.Median follow-up was 40 months (range, 33-44). Twenty-five percent of the patients (n = 4) experienced a DLT within 30 days post-RP; however, the trial was stopped early (n = 16 of planned 38 patients) owing to the proportion and severity of the late adverse events. Post-RP grade 3 genitourinary and gastrointestinal toxicities occurred in 75% (n = 12) and 25% (n = 4) of patients, respectively. Two patients required cystectomy and urinary diversion ≥2 years post-RP. At 24 months post-RP, 75% (n = 12) of men used ≥1 pad/d and 0% had erections suitable for intercourse. Surgical margins were negative in all patients and 31% (n = 5) had complete or partial (pre-RP) MRI-response to SBRT. Three-year biochemical recurrence and distant metastasis were 45% (95% CI, 5%-68%) and 28% (95% CI, 0%-49%), respectively.Neoadjuvant SBRT followed by RP resulted in unacceptably high toxicity and severe QoL declines.

IntroductionProstate cancer (PCa) is the most common cancer in Canadian men. Current models of survivorship care are no longer adequate to address the chronic and complex survivorship needs of patients today. Virtual care models for cancer survivorship have recently been associated with comparable clinical outcomes and lower costs to traditional follow-up care, with patients favouring off-site and on-demand visits. Building on their viability, our research group conceived the Ned Clinic—a virtual PCa survivorship model that provides patients with access to lab results, collects patient-reported outcomes, alerts clinicians to emerging issues, and promotes patient self-care. Despite the promise of the Ned Clinic, the model remains limited by its dependence on oncology specialists, lack of an autonomous triage algorithm, and has only been implemented among PCa survivors living in Ontario.Methods and analysisOur programme of research comprises two main research objectives: (1) to evaluate the process and cost of implementing and sustaining five nurse-led virtual PCa survivorship clinics in three provinces across Canada and identify barriers and facilitators to implementation success and (2) to assess the impact of these virtual clinics on implementation and effectiveness outcomes of enrolled PCa survivors. The design phase will involve developing an autonomous triage algorithm and redesigning the Ned Clinic towards a nurse-led service model. Site-specific implementation plans will be developed to deploy a localised nurse-led virtual clinic at each centre. Effectiveness will be evaluated using a historical control study comparing the survivorship outcomes of 300 PCa survivors enrolled in the Ned Clinic with 300 PCa survivors receiving traditional follow-up care.Ethics and disseminationAppropriate site-specific ethics approval will be secured prior to each research phase. Knowledge translation efforts will include diffusion, dissemination, and application approaches to ensure that knowledge is translated to both academic and lay audiences.

We prospectively investigated the performance of the prostate-specific membrane antigen (PSMA) ligand

There is clear evidence to suggest that diabetes does not affect all populations equally. Among adults living with diabetes, those from ethnoracial minority communities—foreign-born, immigrant, refugee, and culturally marginalized—are at increased risk of poor health outcomes. Artificial intelligence (AI) is actively being researched as a means of improving diabetes management and care; however, several factors may predispose AI to ethnoracial bias. To better understand whether diabetes AI interventions are being designed in an ethnoracially equitable manner, we conducted a secondary analysis of 141 articles included in a 2018 review by Contreras and Vehi entitled “Artificial Intelligence for Diabetes Management and Decision Support: Literature Review.” Two members of our research team independently reviewed each article and selected those reporting ethnoracial data for further analysis. Only 10 articles (7.1%) were ultimately selected for secondary analysis in our case study. Of the 131 excluded articles, 118 (90.1%) failed to mention participants’ ethnic or racial backgrounds. The included articles reported ethnoracial data under various categories, including race (n=6), ethnicity (n=2), race/ethnicity (n=3), and percentage of Caucasian participants (n=1). Among articles specifically reporting race, the average distribution was 69.5% White, 17.1% Black, and 3.7% Asian. Only 2 articles reported inclusion of Native American participants. Given the clear ethnic and racial differences in diabetes biomarkers, prevalence, and outcomes, the inclusion of ethnoracial training data is likely to improve the accuracy of predictive models. Such considerations are imperative in AI-based tools, which are predisposed to negative biases due to their black-box nature and proneness to distributional shift. Based on our findings, we propose a short questionnaire to assess ethnoracial equity in research describing AI-based diabetes interventions. At this unprecedented time in history, AI can either mitigate or exacerbate disparities in health care. Future accounts of the infancy of diabetes AI must reflect our early and decisive action to confront ethnoracial inequities before they are coded into our systems and perpetuate the very biases we aim to eliminate. If we take deliberate and meaningful steps now toward training our algorithms to be ethnoracially inclusive, we can architect innovations in diabetes care that are bound by the diverse fabric of our society.

Introduction: Testicular cancer is the most commonly diagnosed malignancy in young males. Testicular examination is a non-invasive and inexpensive means of detecting testicular cancer at an early stage. In this project, a set of 3D-printed models was developed to facilitate teaching testicular examination and improving understanding of testicular malignancies among patients and medical learners. Methods: Five scrotum models were designed: a control model with healthy testes, and four models containing a healthy testicle and a testicle with an endophytic mass of varying size. The anatomy, texture, and composition of the 3D-printed models were refined using an iterative process between the design team and urologists. The completed models were assessed by six urologists, two urology nurse practitioners, and 32 medical learners. Participants were asked to inspect and palpate each model, and to provide feedback using a five-point Likert scale. Results: Clinicians reported that the models enabled accurate simulation of a testicular examination involving both healthy and pathological testes (x̅=4.3±1.0). They agreed that the models would be useful teaching tools for both medical learners (x̅=4.8±0.5) and patients (x̅=4.8±0.7). Following an educational session with the models, medical learners reported improvements in confidence and skill in performing a testicular examination. Conclusions: 3D-printed models can effectively simulate palpation of both healthy and pathological testes. The developed models have the potential to be a useful adjunct in teaching testicular examination and in demonstrating abnormal findings that require further investigation.

UNSTRUCTURED There is clear evidence to suggest that diabetes does not affect all populations equally. Among adults living with diabetes, those from ethnoracial minority communities—foreign-born, immigrant, refugee, and culturally marginalized—are at increased risk of poor health outcomes. Artificial intelligence (AI) is actively being researched as a means of improving diabetes management and care; however, several factors may predispose AI to ethnoracial bias. To better understand whether diabetes AI interventions are being designed in an ethnoracially equitable manner, we conducted a secondary analysis of 141 articles included in a 2018 review by Contreras and Vehi entitled “Artificial Intelligence for Diabetes Management and Decision Support: Literature Review.” Two members of our research team independently reviewed each article and selected those reporting ethnoracial data for further analysis. Only 10 articles (7.1%) were ultimately selected for secondary analysis in our case study. Of the 131 excluded articles, 118 (90.1%) failed to mention participants’ ethnic or racial backgrounds. The included articles reported ethnoracial data under various categories, including race (n=6), ethnicity (n=2), race/ethnicity (n=3), and percentage of Caucasian participants (n=1). Among articles specifically reporting race, the average distribution was 69.5% White, 17.1% Black, and 3.7% Asian. Only 2 articles reported inclusion of Native American participants. Given the clear ethnic and racial differences in diabetes biomarkers, prevalence, and outcomes, the inclusion of ethnoracial training data is likely to improve the accuracy of predictive models. Such considerations are imperative in AI-based tools, which are predisposed to negative biases due to their black-box nature and proneness to distributional shift. Based on our findings, we propose a short questionnaire to assess ethnoracial equity in research describing AI-based diabetes interventions. At this unprecedented time in history, AI can either mitigate or exacerbate disparities in health care. Future accounts of the infancy of diabetes AI must reflect our early and decisive action to confront ethnoracial inequities before they are coded into our systems and perpetuate the very biases we aim to eliminate. If we take deliberate and meaningful steps now toward training our algorithms to be ethnoracially inclusive, we can architect innovations in diabetes care that are bound by the diverse fabric of our society.

Urothelial carcinoma in a crossed fused renal ectopia (CFRE) is an exceedingly rare clinical finding. We describe the surgical management used to treat upper tract urothelial carcinoma in a 64-year-old man with a right-to-left CFRE. Nephroureterectomy with bladder cuff excision was the treatment of choice. The fused kidney was carefully dissected until an area of demarcation emerged between the vasculature supplying the left and right moieties. Pressure was applied to the isthmus separating the two moieties, and a sharp incision was made to release the left moiety. The operation was completed with limited blood loss. Pathology revealed a high-grade T3 papillary urothelial carcinoma with negative margins. To our knowledge, the case is the first to report urothelial carcinoma in a right-to-left CFRE.

Background The prevalence of non-communicable diseases (NCDs) is rising in low- and middle-income countries (LMICs). Self-management, which enables patients to better manage their health, presents a potentially-scalable means of mitigating the growing burden of NCDs in LMICs. Though the effectiveness of self-management interventions in high-income countries is well-documented, the use of these strategies in LMICs has yet to be thoroughly summarized. Objective The purpose of this scoping review is to summarize the nature and effectiveness of past interventions that have enabled the self-management of NCDs in LMICs. Methods Using the scoping review methodology proposed by Arksey and O'Malley, PubMed was searched for relevant articles published between January 2007 and December 2018. The implemented search strategy comprised three major themes: self-management, NCDs and LMICs. Results Thirty-six original research articles were selected for inclusion. The selected studies largely focused on the self-management of diabetes (N = 21), hypertension (N = 7) and heart failure (N = 5). Most interventions involved the use of short message service (SMS, N = 17) or phone calls (N = 12), while others incorporated educational sessions (N = 10) or the deployment of medical devices (N = 4). The interventions were generally effective and often led to improvements in physiologic indicators, patient self-care and/or patient quality of life. However, the studies emphasized results in small populations, with little indication of future scaling of the intervention. Furthermore, the results indicate a need for further research into the self-management of cardiovascular diseases, as well as for the co-management of diabetes and cardiovascular disease. Conclusions Self-management appears to be an effective means of improving health outcomes in LMICs. Future strategies should include patients and clinicians in all stages of design and development, allowing for a focus on long-term sustainability, scalability and interoperability of the intervention in the target setting.

Background Prognostication of heart failure patients from cardiopulmonary exercise test (CPET) currently involves simplification of complex time-series data into summary indices. We hypothesized that prognostication could be improved by considering the totality of the data generated during a CPET, instead of using summary indices alone. Methods and Results Complete data from 1156 CPETs were used to predict clinical deterioration (characterized by initiation of mechanical circulatory support, listing for heart transplantation or mortality) 1 year post-CPET. We compared the prognostic value (area under the receiver operating characteristic curve) of (1) the most predictive summary indices, (2) staged data collected at discrete intervals using multivariable regression models, and (3) breath-by-breath data using a feedforward neural network. The top-performing models were compared with the commonly used CPET risk score, using absolute net reclassification index. All models were trained and assessed using a 100-iteration Monte Carlo cross-validation. A total of 190 (16.4%) patients experienced clinical deterioration. The summary indices demonstrated subpar discriminative value (area under the receiver operating characteristic curve ≤0.800). Each multivariable model outperformed the summary indices, with the neural network incorporating breath-by-breath data achieving the best performance (area under the receiver operating characteristic curve =0.842). When compared with the CPET risk score (area under the receiver operating characteristic curve =0.759), the top-performing model obtained a net reclassification index of 4.9%. Conclusions The current practice of considering summary indices in isolation fails to realize the full value of CPET data. This may lead to less accurate prognostication of patients and in consequence, inaccurate selection of patients for advanced therapy. Clinical practices, like CPET prognostication, must be continuously reevaluated to ensure optimal usage of valuable (and oft-underutilized) data sources.

Background Prognostication of heart failure patients from cardiopulmonary exercise test (CPET) currently involves simplification of complex time-series data into summary indices. We hypothesized that prognostication could be improved by considering the totality of the data generated during a CPET, instead of using summary indices alone. Methods and Results Complete data from 1156 CPETs were used to predict clinical deterioration (characterized by initiation of mechanical circulatory support, listing for heart transplantation or mortality) 1 year post-CPET. We compared the prognostic value (area under the receiver operating characteristic curve) of (1) the most predictive summary indices, (2) staged data collected at discrete intervals using multivariable regression models, and (3) breath-by-breath data using a feedforward neural network. The top-performing models were compared with the commonly used CPET risk score, using absolute net reclassification index. All models were trained and assessed using a 100-iteration Monte Carlo cross-validation. A total of 190 (16.4%) patients experienced clinical deterioration. The summary indices demonstrated subpar discriminative value (area under the receiver operating characteristic curve ≤0.800). Each multivariable model outperformed the summary indices, with the neural network incorporating breath-by-breath data achieving the best performance (area under the receiver operating characteristic curve =0.842). When compared with the CPET risk score (area under the receiver operating characteristic curve =0.759), the top-performing model obtained a net reclassification index of 4.9%. Conclusions The current practice of considering summary indices in isolation fails to realize the full value of CPET data. This may lead to less accurate prognostication of patients and in consequence, inaccurate selection of patients for advanced therapy. Clinical practices, like CPET prognostication, must be continuously reevaluated to ensure optimal usage of valuable (and oft-underutilized) data sources.

Detailed epidemiologic examination of the distribution of Kawasaki disease (KD) cases could help elucidate the etiology and pathogenesis of this puzzling condition. Location of residence at KD admission was obtained for patients diagnosed in Canada (excluding Quebec) between March 2004 and March 2015. We identified 4,839 patients, 164 of whom (3.4%) developed a coronary artery aneurysm (CAA). A spatiotemporal clustering analysis was performed to determine whether non-random clusters emerged in the distributions of KD and CAA cases. A high-incidence KD cluster occurred in Toronto, ON, between October 2004 and May 2005 (116 cases; relative risk (RR) = 3.43; p

Background: For ethical, legal and administrative reasons, patients have to give explicit consent to a medical treatment. Objectives: This paper identifies the design requirements for electronic treatment consent (eConsent) architecture, and subsequently proposes a model for the eConsent architecture based on the HL7 FHIR® standard. Methods: Six requirements for the eConsent architecture were identified. A conceptual model for the system was then developed to address the identified requirements using HL7 FHIR. Results: The proposed concept makes use of the existing consent model of HL7 FHIR, and includes additional resources for presenting the information to the patient. Moreover, it uses the SNOMED CT terminology to enable semantic interoperability with other health information systems. Conclusions: The proposed eConsent architecture meets the identified requirements. That said, the system is limited by the low maturity of the implemented FHIR resources and the fact that the terminology is currently inexhaustive for the use case. Custom extensions of the used FHIR resources must be considered.

Background The study aims to assess the accuracy of multi-parametric prostate MRI (mpMRI) and 18F-choline PET/CT in tumor segmentation for clinically significant prostate cancer. 18F-choline PET/CT and 3 T mpMRI were performed in 10 prospective subjects prior to prostatectomy. All subjects had a single biopsy-confirmed focus of Gleason ≥ 3+4 cancer. Two radiologists (readers 1 and 2) determined tumor boundaries based on in vivo mpMRI sequences, with clinical and pathologic data available. 18F-choline PET data were co-registered to T2-weighted 3D sequences and a semi-automatic segmentation routine was used to define tumor volumes. Registration of whole-mount surgical pathology to in vivo imaging was conducted utilizing two ex vivo prostate specimen MRIs, followed by gross sectioning of the specimens within a custom-made 3D-printed plastic mold. Overlap and similarity coefficients of manual segmentations (seg1, seg2) and 18F-choline-based segmented lesions (seg3) were compared to the pathologic reference standard. Results All segmentation methods greatly underestimated the true tumor volumes. Human readers (seg1, seg2) and the PET-based segmentation (seg3) underestimated an average of 79, 80, and 58% of the tumor volumes, respectively. Combining segmentation volumes (union of seg1, seg2, seg3 = seg4) decreased the mean underestimated tumor volume to 42% of the true tumor volume. When using the combined segmentation with 5 mm contour expansion, the mean underestimated tumor volume was significantly reduced to 0.03 ± 0.05 mL (2.04 ± 2.84%). Substantial safety margins up to 11–15 mm were needed to include all tumors when the initial segmentation boundaries were drawn by human readers or the semi-automated 18F-choline segmentation tool. Combining MR-based human segmentations with the metabolic information based on 18F-choline PET reduced the necessary safety margin to a maximum of 9 mm to cover all tumors entirely. Conclusions To improve the outcome of focal therapies for significant prostate cancer, it is imperative to recognize the full extent of the underestimation of tumor volumes by mpMRI. Combining metabolic information from 18F-choline with MRI-based segmentation can improve tumor coverage. However, this approach requires confirmation in further clinical studies.

The increasing stress on the global healthcare system driven by the rise of chronic disease and an aging population is necessitating an emphasis on proactive health monitoring and self-management. Potential exists in the wave of emerging wearable devices and the internet of things (IoT) to support a movement towards the decentralization of healthcare. In particular, mobility impairments caused by injury or chronic disease are a major source of concern in the aging population. Individuals with mobility impairments often rely on assistive devices, such as canes or walkers to increase safety and stability. Given the prevalence of assistive devices among these users, instrumenting and connecting assistive technologies could be an effective means of unobtrusive activity monitoring. In this work we present an affordable hybrid sensorized cane, capable of measuring loading, mobility and stability information. The proposed system, which is nearly indistinguishable from a traditional cane, collects these data and then wirelessly transmits them to a mobile device for cloud storage and analysis. A multi-sensor fusion algorithm was used to segment valid gait cycles and identify various temporal gait events. Based on preliminary results, it is believed that the proposed system will be able to identify a variety of gait perturbations, potentially offering future applications in early diagnosis and the management of chronic conditions.

Background Rare hematopoietic stem cell populations are responsible for the transplantation engraftment process. Umbilical cord blood (UCB) is usually processed to the total nucleated cell (TNC), but not to the mononuclear cell (MNC) fraction. TNC counts are used to determine UCB unit storage, release for transplantation and correlation with time to engraftment. However, the TNC fraction contains varying concentrations of red blood cells, granulocytes, platelets and other cells that dilute and mask the stem cells from being detected. This does not allow the quality and potency of the stem cells to be reliably measured. Methods 63 UCB segments and 10 UCB units plus segments were analyzed for the response of both primitive lympho-hematopoietic and primitive hematopoietic stem cells in both the TNC and MNC fractions. The samples were analyzed using a highly sensitive, standardized and validated adenosine triphosphate (ATP) bioluminescence stem cell proliferation assay verified against the colony-forming unit (CFU) assay. Dye exclusion and metabolic viability were also determined. Results Regardless of whether the cells were derived from a segment or unit, the TNC fraction always produced a significantly lower and more variable stem cell response than that derived from the MNC fraction. Routine dye exclusion cell viability did not correspond with metabolic viability and stem cell response. Paired UCB segments produced highly variable results, and the UCB segment did not produce similar results to the unit. Discussion The TNC fraction underestimates the ability and capacity of the stem cells in both the UCB segment and unit and therefore provides an erroneous interpretation of the of the results. Dye exclusion viability can result in false positive values, when in fact the stem cells may be dead or incapable of proliferation. The difference in response between the segment and unit calls into question the ability to use the segment as a representative sample of the UCB unit. It is apparent that present UCB processing and testing methods are inadequate to properly determine the quality and potency of the unit for release and use in a patient.

Background. Chikungunya virus (CHIKV) causes outbreaks of chikungunya fever worldwide and represents an emerging pandemic threat. Vaccine development against CHIKV has proved challenging. Currently there is no approved vaccine or specific therapy for the disease. Methods. To develop novel experimental CHIKV vaccine, we used novel immunization DNA (iDNA) infectious clone technology, which combines the advantages of DNA and live attenuated vaccines. Here we describe an iDNA vaccine composed of plasmid DNA that encode the full-length infectious genome of live attenuated CHIKV clone 181/25 downstream from a eukaryotic promoter. The iDNA approach was designed to initiate replication of live vaccine virus from the plasmid in vitro and in vivo. Results. Experimental CHIKV iDNA vaccines were prepared and evaluated in cultured cells and in mice. Transfection with 10 ng of iDNA was sufficient to initiate replication of vaccine virus in vitro. Vaccination of BALB/c mice with a single 10 μg of CHIKV iDNA plasmid resulted in seroconversion, elicitation of neutralizing antibodies, and protection from experimental challenge with a neurovirulent CHIKV. Conclusions. Live attenuated CHIKV 181/25 vaccine can be delivered in vitro and in vivo by using DNA vaccination. The iDNA approach appears to represent a promising vaccination strategy for CHIK and other alphaviral diseases.

265 THE TOTAL NUCLEATED CELL (TNC) FRACTION AND DYE EXCLUSION VIABILITY CANNOT BE USED TO MEASURE THE QUALITY AND POTENCY OF UMBILICAL CORD BLOOD CELLS IN SEGMENTS AND UNITS IN Rich, J Patterson, CH Moore, E Palser, J Hearn, D Dumitru, HA Harper HemoGenix, Inc, Colorado Springs, Colorado, United States, Beth-Ell College of Nursing and Health Sciences, University of Colorado at Colorado Springs, Colorado Springs, Colorado, United States, Department of Biological Engineering, University of Colorado, Boulder, Colorado, United States, Department of Biology, University of Colorado at Colorado Springs, Colorado Springs, Colorado, United States