An "All-Data-on-Hand" Deep Learning Model to Predict Hospitalization for Diabetic Ketoacidosis in Youth With Type 1 Diabetes: Development and Validation Study.

BACKGROUND: Although prior research has identified multiple risk factors for diabetic ketoacidosis (DKA), clinicians continue to lack clinic-ready models to predict dangerous and costly episodes of DKA. We asked whether we could apply deep learning, specifically the use of a long short-term memory (LSTM) model, to accurately predict the 180-day risk of DKA-related hospitalization for youth with type 1 diabetes (T1D). OBJECTIVE: We aimed to describe the development of an LSTM model to predict the 180-day risk of DKA-related hospitalization for youth with T1D. METHODS: We used 17 consecutive calendar quarters of clinical data (January 10, 2016, to March 18, 2020) for 1745 youths aged 8 to 18 years with T1D from a pediatric diabetes clinic network in the Midwestern United States. The input data included demographics, discrete clinical observations (laboratory results, vital signs, anthropometric measures, diagnosis, and procedure codes), medications, visit counts by type of encounter, number of historic DKA episodes, number of days since last DKA admission, patient-reported outcomes (answers to clinic intake questions), and data features derived from diabetes- and nondiabetes-related clinical notes via natural language processing. We trained the model using input data from quarters 1 to 7 (n=1377), validated it using input from quarters 3 to 9 in a partial out-of-sample (OOS-P; n=1505) cohort, and further validated it in a full out-of-sample (OOS-F; n=354) cohort with input from quarters 10 to 15. RESULTS: DKA admissions occurred at a rate of 5% per 180-days in both out-of-sample cohorts. In the OOS-P and OOS-F cohorts, the median age was 13.7 (IQR 11.3-15.8) years and 13.1 (IQR 10.7-15.5) years; median glycated hemoglobin levels at enrollment were 8.6% (IQR 7.6%-9.8%) and 8.1% (IQR 6.9%-9.5%); recall was 33% (26/80) and 50% (9/18) for the top-ranked 5% of youth with T1D; and 14.15% (213/1505) and 12.7% (45/354) had prior DKA admissions (after the T1D diagnosis), respectively. For lists rank ordered by the probability of hospitalization, precision increased from 33% to 56% to 100% for positions 1 to 80, 1 to 25, and 1 to 10 in the OOS-P cohort and from 50% to 60% to 80% for positions 1 to 18, 1 to 10, and 1 to 5 in the OOS-F cohort, respectively. CONCLUSIONS: The proposed LSTM model for predicting 180-day DKA-related hospitalization was valid in this sample. Future research should evaluate model validity in multiple populations and settings to account for health inequities that may be present in different segments of the population (eg, racially or socioeconomically diverse cohorts). Rank ordering youth by probability of DKA-related hospitalization will allow clinics to identify the most at-risk youth. The clinical implication of this is that clinics may then create and evaluate novel preventive interventions based on available resources.

Diabetes conflict outstrips the positive impact of self-efficacy on youth adherence and glycemic control in type 1 diabetes.

OBJECTIVE: To examine whether self-efficacy buffers the deleterious consequences of diabetes-specific family conflict on self-monitoring blood glucose (SMBG) and glycated hemoglobin (HbA1c) in youth with type 1 diabetes mellitus (T1DM). METHODS: A total of 129 youth with T1DM (aged 10-16 years) completed measures of diabetes-specific family conflict and self-efficacy for diabetes management, and their blood glucose meter data and HbA1c were extracted from the electronic medical record. We preformed moderation analyses to examine whether self-efficacy moderated the association that diabetes-specific family conflict had with SMBG and HbA1c. We used simple slopes analyses to probe significant interactions. RESULTS: Our results indicated that self-efficacy moderated the association that diabetes-specific family conflict had with SMBG and HbA1c. The pattern of these findings showed that high self-efficacy buffered the negative impact of diabetes conflict on HbA1c. However, benefits of high self-efficacy for more frequent SMBG was only apparent in the context of low diabetes-specific family conflict. CONCLUSIONS: Study findings highlight the interactive relationship between diabetes-specific family conflict and self-efficacy in relation to SMBG and glycemic control. These findings suggest that family functioning and youth's self-efficacy are promising intervention targets for families having trouble with SMBG and HbA1c.

Automated video-based assessment of surgical skills for training and evaluation in medical schools.

PURPOSE: Routine evaluation of basic surgical skills in medical schools requires considerable time and effort from supervising faculty. For each surgical trainee, a supervisor has to observe the trainees in person. Alternatively, supervisors may use training videos, which reduces some of the logistical overhead. All these approaches however are still incredibly time consuming and involve human bias. In this paper, we present an automated system for surgical skills assessment by analyzing video data of surgical activities. METHOD: We compare different techniques for video-based surgical skill evaluation. We use techniques that capture the motion information at a coarser granularity using symbols or words, extract motion dynamics using textural patterns in a frame kernel matrix, and analyze fine-grained motion information using frequency analysis. RESULTS: We were successfully able to classify surgeons into different skill levels with high accuracy. Our results indicate that fine-grained analysis of motion dynamics via frequency analysis is most effective in capturing the skill relevant information in surgical videos. CONCLUSION: Our evaluations show that frequency features perform better than motion texture features, which in-turn perform better than symbol-/word-based features. Put succinctly, skill classification accuracy is positively correlated with motion granularity as demonstrated by our results on two challenging video datasets.

Hemoglobin A1c (HbA1c) changes over time among adolescent and young adult participants in the T1D exchange clinic registry.

OBJECTIVE: Hemoglobin A1c (HbA1c) levels among individuals with type 1 diabetes (T1D) influence the longitudinal risk for diabetes-related complications. Few studies have examined HbA1c trends across time in children, adolescents, and young adults with T1D. This study examines changes in glycemic control across the specific transition periods of pre-adolescence-to-adolescence and adolescence-to-young adulthood, and the demographic and clinical factors associated with these changes. RESEARCH DESIGN AND METHODS: Available HbA1c lab results for up to 10 yr were collected from medical records at 67 T1D Exchange clinics. Two retrospective cohorts were evaluated: the pre-adolescent-to-adolescent cohort consisting of 85 016 HbA1c measurements from 6574 participants collected when the participants were 8-18 yr old and the adolescent-to-young adult cohort, 2200 participants who were 16-26 yr old at the time of 17 279 HbA1c measurements. RESULTS: HbA1c in the 8-18 cohort increased over time after age 10 yr until ages 16-17; followed by a plateau. HbA1c levels in the 16-26 cohort remained steady from 16-18, and then gradually declined. For both cohorts, race/ethnicity, income, health insurance, and pump use were all significant in explaining individual variations in age-centered HbA1c (p < 0.001). For the 8-18 cohort, insulin pump use, age of onset, and health insurance were significant in predicting individual HbA1c trajectory. CONCLUSIONS: Glycemic control among patients 8-18 yr old worsens over time, through age 16. Elevated HbA1c levels observed in 18 yr-olds begin a steady improvement into early adulthood. Focused interventions to prevent deterioration in glucose control in pre-adolescence, adolescence, and early adulthood are needed.

The Cost of a Healthier Diet for Young Children With Type 1 Diabetes Mellitus.

OBJECTIVE: This study used a market-basket approach to examine the availability and cost of a standard food shopping list (R-TFP) vs a healthier food shopping list (H-TFP) in the grocery stores used by a sample of 23 families of young children with type 1 diabetes mellitus (T1DM). METHODS: The researchers used frequency counts to measure availability. The average cost of the R-TFP and H-TFP was compared using paired t test. RESULTS: Small or independent markets had the highest percentage of missing foods (14%), followed by chain supermarkets (3%) and big box stores (2%). There was a significant difference in average cost for the R-TFP vs the H-TFP ($324.71 and $380.07, respectively; P < .001). CONCLUSIONS AND IMPLICATIONS: Families may encounter problems finding healthier foods and/or incur greater costs for healthier foods. Nutrition education programs for T1DM need to teach problem solving to help families overcome these barriers.