IU Indianapolis ScholarWorks :: Browsing by Subject "Prediction"

Browsing by Subject "Prediction"

Now showing 1 - 10 of 28

Assessing Risk of Future Suicidality in Emergency Department Patients
(Wiley, 2020-04-02) Brucker, Krista; Duggan, Carter; Niezer, Joseph; Roseberry, Kyle; Le-Niculescu, Helen; Niculescu, Alexander B.; Kline, Jeffrey A.; Emergency Medicine, School of Medicine
Background. Emergency Departments (ED) are the first line of evaluation for patients at risk and in crisis, with or without overt suicidality (ideation, attempts). Currently employed triage and assessments methods miss some of the individuals who subsequently become suicidal. The Convergent Functional Information for Suicidality (CFI-S) 22 item checklist of risk factors, that does not ask directly about suicidal ideation, has demonstrated good predictive ability for suicidality in previous studies in psychiatric outpatients, but has not been tested in the real world-setting of emergency departments (EDs). Methods. We administered CFI-S prospectively to a convenience sample of consecutive ED patients. Median administration time was 3 minutes. Patients were also asked at triage about suicidal thoughts or intentions per standard ED suicide clinical screening (SCS), and the treating ED physician was asked to fill a physician gestalt visual analog scale (VAS) for likelihood of future suicidality spectrum events (SSE) (ideation, preparatory acts, attempts, completed suicide). We performed structured chart review and telephone follow-up at 6 months post index visit. Results. The median time to complete the CFI-S was three minutes (1st to 3rd quartile 3–6 minutes). Of the 338 patients enrolled, 45 (13.3%) were positive on the initial SCS, and 32 (9.5%) experienced a SSE in the 6 months follow-up. Overall, across genders, SCS had a modest diagnostic discrimination for future SSE (ROC AUC 0.63,). The physician VAS was better (AUC 0.76 CI 0.66–0.85), and the CFI-S was slightly higher (AUC 0.81, CI 0.76–0.87). The top CFI-S differentiating items were psychiatric illness, perceived uselessness, and social isolation. The top CFI-S items were family history of suicide, age, and past history of suicidal acts. Conclusions. Using CFI-S, or some of its items, in busy EDs may help improve the detection of patients at high risk for future suicidality.
Can Early-Assignment Grades Predict Final Grades in IT Courses?: American Society for Engineering Education
(2017) Ramanathan, Parameswari; Fernandez, Eugenia; Computer Information and Graphics Technology, School of Engineering and Technology
Classification and prediction of cognitive trajectories of cognitively unimpaired individuals
(Frontiers Media, 2023-03-13) Kim, Young Ju; Kim, Si Eun; Hahn, Alice; Jang, Hyemin; Kim, Jun Pyo; Kim, Hee Jin; Na, Duk L.; Chin, Juhee; Seo, Sang Won; Alzheimer’s Disease Neuroimaging Initiative; Radiology and Imaging Sciences, School of Medicine
Objectives: Efforts to prevent Alzheimer's disease (AD) would benefit from identifying cognitively unimpaired (CU) individuals who are liable to progress to cognitive impairment. Therefore, we aimed to develop a model to predict cognitive decline among CU individuals in two independent cohorts. Methods: A total of 407 CU individuals from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and 285 CU individuals from the Samsung Medical Center (SMC) were recruited in this study. We assessed cognitive outcomes by using neuropsychological composite scores in the ADNI and SMC cohorts. We performed latent growth mixture modeling and developed the predictive model. Results: Growth mixture modeling identified 13.8 and 13.0% of CU individuals in the ADNI and SMC cohorts, respectively, as the "declining group." In the ADNI cohort, multivariable logistic regression modeling showed that increased amyloid-β (Aβ) uptake (β [SE]: 4.852 [0.862], p < 0.001), low baseline cognitive composite scores (β [SE]: -0.274 [0.070], p < 0.001), and reduced hippocampal volume (β [SE]: -0.952 [0.302], p = 0.002) were predictive of cognitive decline. In the SMC cohort, increased Aβ uptake (β [SE]: 2.007 [0.549], p < 0.001) and low baseline cognitive composite scores (β [SE]: -4.464 [0.758], p < 0.001) predicted cognitive decline. Finally, predictive models of cognitive decline showed good to excellent discrimination and calibration capabilities (C-statistic = 0.85 for the ADNI model and 0.94 for the SMC model). Conclusion: Our study provides novel insights into the cognitive trajectories of CU individuals. Furthermore, the predictive model can facilitate the classification of CU individuals in future primary prevention trials.
Comparisons of Metabolic Measures to Predict T1D vs Detect a Preventive Treatment Effect in High-Risk Individuals
(Oxford University Press, 2024) Sims, Emily K.; Cuthbertson, David; Jacobsen, Laura; Ismail, Heba M.; Nathan, Brandon M.; Herold, Kevan C.; Redondo, Maria J.; Sosenko, Jay; Pediatrics, School of Medicine
Context: Metabolic measures are frequently used to predict type 1 diabetes (T1D) and to understand effects of disease-modifying therapies. Objective: Compare metabolic endpoints for their ability to detect preventive treatment effects and predict T1D. Methods: Six-month changes in metabolic endpoints were assessed for (1) detecting treatment effects by comparing placebo and treatment arms from the randomized controlled teplizumab prevention trial, a multicenter clinical trial investigating 14-day intravenous teplizumab infusion and (2) predicting T1D in the TrialNet Pathway to Prevention natural history study. For each metabolic measure, t-Values from t tests for detecting a treatment effect were compared with chi-square values from proportional hazards regression for predicting T1D. Participants in the teplizumab prevention trial and participants in the Pathway to Prevention study selected with the same inclusion criteria used for the teplizumab trial were studied. Results: Six-month changes in glucose-based endpoints predicted diabetes better than C-peptide-based endpoints, yet the latter were better at detecting a teplizumab effect. Combined measures of glucose and C-peptide were more balanced than measures of glucose alone or C-peptide alone for predicting diabetes and detecting a teplizumab effect. Conclusion: The capacity of a metabolic endpoint to detect a treatment effect does not necessarily correspond to its accuracy for predicting T1D. However, combined glucose and C-peptide endpoints appear to be effective for both predicting diabetes and detecting a response to immunotherapy. These findings suggest that combined glucose and C-peptide endpoints should be incorporated into the design of future T1D prevention trials.
Delirium Severity Trajectories and Outcomes in ICU Patients. Defining a Dynamic Symptom Phenotype
(American Thoracic Society, 2020-09) Lindroth, Heidi; Khan, Babar A.; Carpenter, Janet S.; Gao, Sujuan; Perkins, Anthony J.; Khan, Sikandar H.; Wang, Sophia; Jones, Richard N.; Boustani, Malaz A.; School of Nursing
Rationale: Delirium severity and duration are independently associated with higher mortality and morbidity. No studies to date have described a delirium trajectory by integrating both severity and duration. Objectives: The primary aim was to develop delirium trajectories by integrating symptom severity and duration. The secondary aim was to investigate the association among trajectory membership, clinical characteristics, and 30-day mortality. Methods: A secondary analysis of the PMD (Pharmacologic Management of Delirium) randomized control trial (ClinicalTrials.gov Identifier: NCT00842608; N = 531) was conducted. The presence of delirium and symptom severity were measured at least daily for 7 days using the Confusion Assessment Method for the intensive care unit (CAM-ICU) and CAM-ICU-7 (on a scale of 0-7, with 7 being the most severe). Delirium trajectories were defined using an innovative, data-driven statistical method (group-based trajectory modeling [GBTM]) and SAS v9.4.Results: A total of 531 delirious participants (mean age 60 yr [standard deviation = 16], 55% female, and 46% African American) were analyzed. Five distinct delirium trajectories were described (CAM-ICU-7: mean [standard deviation]); mild-brief (CAM-ICU-7: 0.5 [0.5]), severe-rapid recovers (CAM-ICU-7: 2.1 [1.0]), mild-accelerating (CAM-ICU-7: 2.2 [0.9]), severe-slow recovers (CAM-ICU-7: 3.9 [0.9]), and severe-nonrecovers (CAM-ICU-7: 5.9 [1.0]). Baseline cognition and race were associated with trajectory membership. Trajectory membership independently predicted 30-day mortality while controlling for age, sex, race, cognition, illness severity, and comorbidities. Conclusions: This secondary analysis described five distinct delirium trajectories based on delirium symptom severity and duration using group-based trajectory modeling. Trajectory membership predicted 30-day mortality.
Development and Validation of a Routine Electronic Health Record-Based Delirium Prediction Model for Surgical Patients Without Dementia: Retrospective Case-Control Study
(JMIR, 2025-01-09) Holler, Emma; Ludema, Christina; Ben Miled, Zina; Rosenberg, Molly; Kalbaugh, Corey; Boustani, Malaz; Mohanty, Sanjay; Surgery, School of Medicine
Background: Postoperative delirium (POD) is a common complication after major surgery and is associated with poor outcomes in older adults. Early identification of patients at high risk of POD can enable targeted prevention efforts. However, existing POD prediction models require inpatient data collected during the hospital stay, which delays predictions and limits scalability. Objective: This study aimed to develop and externally validate a machine learning-based prediction model for POD using routine electronic health record (EHR) data. Methods: We identified all surgical encounters from 2014 to 2021 for patients aged 50 years and older who underwent an operation requiring general anesthesia, with a length of stay of at least 1 day at 3 Indiana hospitals. Patients with preexisting dementia or mild cognitive impairment were excluded. POD was identified using Confusion Assessment Method records and delirium International Classification of Diseases (ICD) codes. Controls without delirium or nurse-documented confusion were matched to cases by age, sex, race, and year of admission. We trained logistic regression, random forest, extreme gradient boosting (XGB), and neural network models to predict POD using 143 features derived from routine EHR data available at the time of hospital admission. Separate models were developed for each hospital using surveillance periods of 3 months, 6 months, and 1 year before admission. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). Each model was internally validated using holdout data and externally validated using data from the other 2 hospitals. Calibration was assessed using calibration curves. Results: The study cohort included 7167 delirium cases and 7167 matched controls. XGB outperformed all other classifiers. AUROCs were highest for XGB models trained on 12 months of preadmission data. The best-performing XGB model achieved a mean AUROC of 0.79 (SD 0.01) on the holdout set, which decreased to 0.69-0.74 (SD 0.02) when externally validated on data from other hospitals. Conclusions: Our routine EHR-based POD prediction models demonstrated good predictive ability using a limited set of preadmission and surgical variables, though their generalizability was limited. The proposed models could be used as a scalable, automated screening tool to identify patients at high risk of POD at the time of hospital admission.
Development and Validation of Primary Graft Dysfunction Predictive Algorithm for Lung Transplant Candidates
(Elsevier, 2024) Diamond, Joshua M.; Anderson, Michaela R.; Cantu, Edward; Clausen, Emily S.; Shashaty, Michael G. S.; Kalman, Laurel; Oyster, Michelle; Crespo, Maria M.; Bermudez, Christian A.; Benvenuto, Luke; Palmer, Scott M.; Snyder, Laurie D.; Hartwig, Matthew G.; Wille, Keith; Hage, Chadi; McDyer, John F.; Merlo, Christian A.; Shah, Pali D.; Orens, Jonathan B.; Dhillon, Ghundeep S.; Lama, Vibha N.; Patel, Mrunal G.; Singer, Jonathan P.; Hachem, Ramsey R.; Michelson, Andrew P.; Hsu, Jesse; Localio, A. Russell; Christie, Jason D.; Medicine, School of Medicine
Background: Primary graft dysfunction (PGD) is the leading cause of early morbidity and mortality after lung transplantation. Accurate prediction of PGD risk could inform donor approaches and perioperative care planning. We sought to develop a clinically useful, generalizable PGD prediction model to aid in transplant decision-making. Methods: We derived a predictive model in a prospective cohort study of subjects from 2012 to 2018, followed by a single-center external validation. We used regularized (lasso) logistic regression to evaluate the predictive ability of clinically available PGD predictors and developed a user interface for clinical application. Using decision curve analysis, we quantified the net benefit of the model across a range of PGD risk thresholds and assessed model calibration and discrimination. Results: The PGD predictive model included distance from donor hospital to recipient transplant center, recipient age, predicted total lung capacity, lung allocation score (LAS), body mass index, pulmonary artery mean pressure, sex, and indication for transplant; donor age, sex, mechanism of death, and donor smoking status; and interaction terms for LAS and donor distance. The interface allows for real-time assessment of PGD risk for any donor/recipient combination. The model offers decision-making net benefit in the PGD risk range of 10% to 75% in the derivation centers and 2% to 10% in the validation cohort, a range incorporating the incidence in that cohort. Conclusion: We developed a clinically useful PGD predictive algorithm across a range of PGD risk thresholds to support transplant decision-making, posttransplant care, and enrich samples for PGD treatment trials.
Differential Learning for Outliers: A Case Study of Water Demand Prediction
(MDPI, 2018-11) Shah, Setu; Ben Miled, Zina; Schaefer, Rebecca; Berube, Steve; Electrical and Computer Engineering, School of Engineering and Technology
Predicting water demands is becoming increasingly critical because of the scarcity of this natural resource. In fact, the subject was the focus of numerous studies by a large number of researchers around the world. Several models have been proposed that are able to predict water demands using both statistical and machine learning techniques. These models have successfully identified features that can impact water demand trends for rural and metropolitan areas. However, while the above models, including recurrent network models proposed by the authors are able to predict normal water demands, most have difficulty estimating potential deviations from the norms. Outliers in water demand can be due to various reasons including high temperatures and voluntary or mandatory consumption restrictions by the water utility companies. Estimating these deviations is necessary, especially for water utility companies with a small service footprint, in order to efficiently plan water distribution. This paper proposes a differential learning model that can help model both over-consumption and under-consumption. The proposed differential model builds on a previously proposed recurrent neural network model that was successfully used to predict water demand in central Indiana.
External validation of the modified sepsis renal angina index for prediction of severe acute kidney injury in children with septic shock
(Springer Nature, 2023-11-28) Stanski, Natalja L.; Basu, Rajit K.; Cvijanovich, Natalie Z.; Fitzgerald, Julie C.; Bigham, Michael T.; Jain, Parag N.; Schwarz, Adam J.; Lutfi, Riad; Thomas, Neal J.; Baines, Torrey; Haileselassie, Bereketeab; Weiss, Scott L.; Atreya, Mihir R.; Lautz, Andrew J.; Zingarelli, Basilia; Standage, Stephen W.; Kaplan, Jennifer; Chawla, Lakhmir S.; Goldstein, Stuart L.; Pediatrics, School of Medicine
Background: Acute kidney injury (AKI) occurs commonly in pediatric septic shock and increases morbidity and mortality. Early identification of high-risk patients can facilitate targeted intervention to improve outcomes. We previously modified the renal angina index (RAI), a validated AKI prediction tool, to improve specificity in this population (sRAI). Here, we prospectively assess sRAI performance in a separate cohort. Methods: A secondary analysis of a prospective, multicenter, observational study of children with septic shock admitted to the pediatric intensive care unit from 1/2019 to 12/2022. The primary outcome was severe AKI (≥ KDIGO Stage 2) on Day 3 (D3 severe AKI), and we compared predictive performance of the sRAI (calculated on Day 1) to the original RAI and serum creatinine elevation above baseline (D1 SCr > Baseline +). Original renal angina fulfillment (RAI +) was defined as RAI ≥ 8; sepsis renal angina fulfillment (sRAI +) was defined as RAI ≥ 20 or RAI 8 to < 20 with platelets < 150 × 103/µL. Results: Among 363 patients, 79 (22%) developed D3 severe AKI. One hundred forty (39%) were sRAI + , 195 (54%) RAI + , and 253 (70%) D1 SCr > Baseline + . Compared to sRAI-, sRAI + had higher risk of D3 severe AKI (RR 8.9, 95%CI 5-16, p < 0.001), kidney replacement therapy (KRT) (RR 18, 95%CI 6.6-49, p < 0.001), and mortality (RR 2.5, 95%CI 1.2-5.5, p = 0.013). sRAI predicted D3 severe AKI with an AUROC of 0.86 (95%CI 0.82-0.90), with greater specificity (74%) than D1 SCr > Baseline (36%) and RAI + (58%). On multivariable regression, sRAI + retained associations with D3 severe AKI (aOR 4.5, 95%CI 2.0-10.2, p < 0.001) and need for KRT (aOR 5.6, 95%CI 1.5-21.5, p = 0.01). Conclusions: Prediction of severe AKI in pediatric septic shock is important to improve outcomes, allocate resources, and inform enrollment in clinical trials examining potential disease-modifying therapies. The sRAI affords more accurate and specific prediction than context-free SCr elevation or the original RAI in this population.
IA-2A positivity increases risk of progression within and across established stages of type 1 diabetes
(Springer, 2025) Sims, Emily K.; Cuthbertson, David; Ferrat, Lauric A.; Bosi, Emanuele; Evans‑Molina, Carmella; DiMeglio, Linda A.; Nathan, Brandon M.; Ismail, Heba M.; Jacobsen, Laura M.; Redondo, Maria J.; Oram, Richard A.; Sosenko, Jay M.; Pediatrics, School of Medicine
Aims/hypothesis: Accurate understanding of type 1 diabetes risk is critical for optimisation of counselling, monitoring and interventions, yet even within established staging classifications, individual time to clinical disease varies. Previous work has associated IA-2A positivity with increased type 1 diabetes progression but a comprehensive assessment of the impact of screening for IA-2A positivity across the natural history of autoantibody positivity has not been performed. We asked whether IA-2A would consistently be associated with higher risk of progression within and across established stages of type 1 diabetes in a large natural history study. Methods: Genetic, autoantibody and metabolic data from adult and paediatric autoantibody-negative (n=192) and autoantibody-positive (n=4577) relatives of individuals with type 1 diabetes followed longitudinally in the Type 1 Diabetes TrialNet Pathway to Prevention Study were analysed. Cox regression was used to compare cumulative incidences of clinical diabetes by autoantibody profiles and disease stages. Results: Compared with IA-2A- individuals, IA-2A+ individuals had higher genetic risk scores and clinical progression risk within single-autoantibody-positive (5.3-fold increased 5 year risk), stage 1 (2.2-fold increased 5 year risk) and stage 2 (1.3-fold increased 5 year risk) type 1 diabetes categories. Individuals with single-autoantibody positivity for IA-2A showed increased metabolic dysfunction and diabetes progression compared with people who were autoantibody negative, those positive for another single autoantibody, and IA-2A- stage 1 individuals. Individuals at highest risk within the single-IA-2A+ category included children (HR 14.2 [95% CI 1.9, 103.1], p=0.009), individuals with IA-2A titres above the median (HR 3.5 [95% CI 1.9, 6.6], p<0.001), individuals with high genetic risk scores (HR 1.4 [95% CI 1.2,1.6], p<0.001) and individuals with HLA DR4-positive status (HR 3.7 [95% CI 1.6, 8.3], p=0.002). When considering all autoantibody-positive individuals, progression risk was similar for euglycaemic IA-2A+ individuals and dysglycaemic IA-2A- individuals. Conclusions/interpretation: IA-2A positivity is consistently associated with increased progression risk throughout the natural history of type 1 diabetes development. Individuals with single-autoantibody positivity for IA-2A have a greater risk of disease progression than those who meet stage 1 criteria but who are IA-2A-. Approaches to incorporate IA-2A+ status into monitoring strategies for autoantibody-positive individuals should be considered.

Browsing by Subject "Prediction"

Results Per Page

Sort Options