Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

Machine Learning for Predicting Micro- and Macrovascular Complications in Individuals With Prediabetes or Diabetes: Retrospective Cohort Study.

In Journal of medical Internet research ; h5-index 88.0

BACKGROUND : Micro- and macrovascular complications are a major burden for individuals with diabetes and can already arise in a prediabetic state. To allocate effective treatments and to possibly prevent these complications, identification of those at risk is essential.

OBJECTIVE : This study aimed to build machine learning (ML) models that predict the risk of developing a micro- or macrovascular complication in individuals with prediabetes or diabetes.

METHODS : In this study, we used electronic health records from Israel that contain information about demographics, biomarkers, medications, and disease codes; span from 2003 to 2013; and were queried to identify individuals with prediabetes or diabetes in 2008. Subsequently, we aimed to predict which of these individuals developed a micro- or macrovascular complication within the next 5 years. We included 3 microvascular complications: retinopathy, nephropathy, and neuropathy. In addition, we considered 3 macrovascular complications: peripheral vascular disease (PVD), cerebrovascular disease (CeVD), and cardiovascular disease (CVD). Complications were identified via disease codes, and, for nephropathy, the estimated glomerular filtration rate and albuminuria were considered additionally. Inclusion criteria were complete information on age and sex and on disease codes (or measurements of estimated glomerular filtration rate and albuminuria for nephropathy) until 2013 to account for patient dropout. Exclusion criteria for predicting a complication were diagnosis of this specific complication before or in 2008. In total, 105 predictors from demographics, biomarkers, medications, and disease codes were used to build the ML models. We compared 2 ML models: logistic regression and gradient-boosted decision trees (GBDTs). To explain the predictions of the GBDTs, we calculated Shapley additive explanations values.

RESULTS : Overall, 13,904 and 4259 individuals with prediabetes and diabetes, respectively, were identified in our underlying data set. For individuals with prediabetes, the areas under the receiver operating characteristic curve for logistic regression and GBDTs were, respectively, 0.657 and 0.681 (retinopathy), 0.807 and 0.815 (nephropathy), 0.727 and 0.706 (neuropathy), 0.730 and 0.727 (PVD), 0.687 and 0.693 (CeVD), and 0.707 and 0.705 (CVD); for individuals with diabetes, the areas under the receiver operating characteristic curve were, respectively, 0.673 and 0.726 (retinopathy), 0.763 and 0.775 (nephropathy), 0.745 and 0.771 (neuropathy), 0.698 and 0.715 (PVD), 0.651 and 0.646 (CeVD), and 0.686 and 0.680 (CVD). Overall, the prediction performance is comparable for logistic regression and GBDTs. The Shapley additive explanations values showed that increased levels of blood glucose, glycated hemoglobin, and serum creatinine are risk factors for microvascular complications. Age and hypertension were associated with an elevated risk for macrovascular complications.

CONCLUSIONS : Our ML models allow for an identification of individuals with prediabetes or diabetes who are at increased risk of developing micro- or macrovascular complications. The prediction performance varied across complications and target populations but was in an acceptable range for most prediction tasks.

Schallmoser Simon, Zueger Thomas, Kraus Mathias, Saar-Tsechansky Maytal, Stettler Christoph, Feuerriegel Stefan

2023-Feb-27

diabetes, machine learning, macrovascular complications, microvascular complications, prediabetes

Internal Medicine Internal Medicine

Developing supervised machine learning algorithms to evaluate the therapeutic effect and laboratory-related adverse events of cyclosporine and tacrolimus in renal transplants.

In International journal of clinical pharmacy

BACKGROUND : Single nucleotide polymorphisms influence the effects of tacrolimus and cyclosporine in renal transplants.

AIM : We set out to use machine learning algorithms (MLAs) to identify variables that predict the therapeutic effects and adverse events following tacrolimus and cyclosporine administration in renal transplant patients.

METHOD : We sampled 120 adult renal transplant patients (on cyclosporine or tacrolimus). Generalized linear model (GLM), support vector machine (SVM), artificial neural network (ANN), Chi-square automatic interaction detection, classification and regression tree, and K-nearest neighbors were the chosen MLAs. The mean absolute error (MAE), relative mean square error (RMSE), and regression coefficient (β) with a 95% confidence interval (CI) were used as the model parameters.

RESULTS : For a stable dose of tacrolimus, the MAEs (RMSEs) of GLM, SVM, and ANN were 1.3 (1.5), 1.3 (1.8), and 1.7 (2.3) mg/day, respectively. GLM revealed that the POR*28 genotype and age significantly predicted the stable dose of tacrolimus as follows: POR*28 (β -1.8; 95% CI -3, -0.5; p = 0.006), and age (β -0.04; 95% CI -0.1, -0.006; p = 0.02). For a stable dose of cyclosporine, MAEs (RMSEs) of 93.2 (103.4), 79.1 (115.2), and 73.7 (91.7) mg/day were observed with GLM, SVM, and ANN, respectively. GLM revealed the following predictors of a stable dose of cyclosporine: CYP3A5*3 (β -80.8; 95% CI -130.3, -31.2; p = 0.001), and age (β -3.4; 95% CI -5.9, -0.9; p = 0.007).

CONCLUSION : We observed that various MLAs could identify significant predictors that were useful to optimize tacrolimus and cyclosporine dosing regimens; yet, the findings must be externally validated.

Sridharan Kannan, Shah Shamik

2023-Feb-27

Artificial intelligence, Cyclosporine, Immunosuppressants, Machine learning algorithms, Tacrolimus

General General

Automatic vertebral fracture and three-column injury diagnosis with fracture visualization by a multi-scale attention-guided network.

In Medical & biological engineering & computing ; h5-index 32.0

Deep learning methods have the potential to improve the efficiency of diagnosis for vertebral fractures with computed tomography (CT) images. Most existing intelligent vertebral fracture diagnosis methods only provide dichotomized results at a patient level. However, a fine-grained and more nuanced outcome is clinically needed. This study proposed a novel network, a multi-scale attention-guided network (MAGNet), to diagnose vertebral fractures and three-column injuries with fracture visualization at a vertebra level. By imposing attention constraints through a disease attention map (DAM), a fusion of multi-scale spatial attention maps, the MAGNet can get task highly relevant features and localize fractures. A total of 989 vertebrae were studied here. After four-fold cross-validation, the area under the ROC curve (AUC) of our model for vertebral fracture dichotomized diagnosis and three-column injury diagnosis was 0.884 ± 0.015 and 0.920 ± 0.104, respectively. The overall performance of our model outperformed classical classification models, attention models, visual explanation methods, and attention-guided methods based on class activation mapping. Our work can promote the clinical application of deep learning to diagnose vertebral fractures and provide a way to visualize and improve the diagnosis results with attention constraints.

Zhang Shunan, Zhao Ziqi, Qiu Lu, Liang Duan, Wang Kun, Xu Jun, Zhao Jun, Sun Jianqi

2023-Feb-27

Deep learning, Disease attention map, Fracture visualization, Three-column injury, Vertebral fracture diagnosis

General General

Prediction of gestational diabetes using deep learning and Bayesian optimization and traditional machine learning techniques.

In Medical & biological engineering & computing ; h5-index 32.0

The study aimed to develop a clinical diagnosis system to identify patients in the GD risk group and reduce unnecessary oral glucose tolerance test (OGTT) applications for pregnant women who are not in the GD risk group using deep learning algorithms. With this aim, a prospective study was designed and the data was taken from 489 patients between the years 2019 and 2021, and informed consent was obtained. The clinical decision support system for the diagnosis of GD was developed using the generated dataset with deep learning algorithms and Bayesian optimization. As a result, a novel successful decision support model was developed using RNN-LSTM with Bayesian optimization that gave 95% sensitivity and 99% specificity on the dataset for the diagnosis of patients in the GD risk group by obtaining 98% AUC (95% CI (0.95-1.00) and p < 0.001). Thus, with the clinical diagnosis system developed to assist physicians, it is planned to save both cost and time, and reduce possible adverse effects by preventing unnecessary OGTT for patients who are not in the GD risk group.

Kurt Burçin, Gürlek Beril, Keskin Seda, Özdemir Sinem, Karadeniz Özlem, Kırkbir İlknur Buçan, Kurt Tuğba, Ünsal Serbülent, Kart Cavit, Baki Neslihan, Turhan Kemal

2023-Feb-27

Bayesian optimization, Clinical decision support system, Deep learning, Gestational diabetes (GD), Random forest, SVM

Radiology Radiology

Deep learning image reconstruction algorithm: impact on image quality in coronary computed tomography angiography.

In La Radiologia medica

PURPOSE : To perform a comprehensive intraindividual objective and subjective image quality evaluation of coronary CT angiography (CCTA) reconstructed with deep learning image reconstruction (DLIR) and to assess correlation with routinely applied hybrid iterative reconstruction algorithm (ASiR-V).

MATERIAL AND METHODS : Fifty-one patients (29 males) undergoing clinically indicated CCTA from April to December 2021 were prospectively enrolled. Fourteen datasets were reconstructed for each patient: three DLIR strength levels (DLIR_L, DLIR_M, and DLIR_H), ASiR-V from 10% to 100% in 10%-increment, and filtered back-projection (FBP). Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) determined objective image quality. Subjective image quality was assessed with a 4-point Likert scale. Concordance between reconstruction algorithms was assessed by Pearson correlation coefficient.

RESULTS : DLIR algorithm did not impact vascular attenuation (P ≥ 0.374). DLIR_H showed the lowest noise, comparable with ASiR-V 100% (P = 1) and significantly lower than other reconstructions (P ≤ 0.021). DLIR_H achieved the highest objective quality, with SNR and CNR comparable to ASiR-V 100% (P = 0.139 and 0.075, respectively). DLIR_M obtained comparable objective image quality with ASiR-V 80% and 90% (P ≥ 0.281), while achieved the highest subjective image quality (4, IQR: 4-4; P ≤ 0.001). DLIR and ASiR-V datasets returned a very strong correlation in the assessment of CAD (r = 0.874, P = 0.001).

CONCLUSION : DLIR_M significantly improves CCTA image quality and has very strong correlation with routinely applied ASiR-V 50% dataset in the diagnosis of CAD.

De Santis Domenico, Polidori Tiziano, Tremamunno Giuseppe, Rucci Carlotta, Piccinni Giulia, Zerunian Marta, Pugliese Luca, Del Gaudio Antonella, Guido Gisella, Barbato Luca, Laghi Andrea, Caruso Damiano

2023-Feb-27

Artificial intelligence, CCTA, Coronary computed tomography angiography, Deep learning, Image reconstruction

Surgery Surgery

Proximity to High/Moderate vs Low Diversity Selection Food Stores and Patient Weight loss through 24 Months.

In Obesity surgery ; h5-index 57.0

PURPOSE : Explorations into the neighborhood food environment have not adequately extended to adults with obesity who undergo bariatric surgery. The objective of this study is to determine how diversity of food selection at food retail stores within proximities of 5- and 10-min walks associate with patient postoperative weight loss over 24 months.

MATERIALS AND METHODS : Eight hundred eleven patients (82.1% female; 60.0% White) who had primary bariatric surgery (48.6% gastric bypass) from 2015 to 2019 at The Ohio State University were included. EHR variables included race, insurance, procedure, and percent total weight loss (%TWL) at 2, 3, 6, 12, and 24 months. Proximity from patients' home addresses to food stores within a 5- (0.25 mile)- and 10-min (0.50 mile) walk were totaled for low (LD) and moderate/high (M/HD) diversity food selections. Bivariate analyses were conducted with %TWL at all visits and LD and M/HD selections within 5- (0, ≥ 1) and 10-min (0, 1, ≥ 2) walk proximities. Four mixed multilevel models were conducted with dependent variable %TWL over 24 months with visits as the between subjects factor and covariates: race, insurance, procedure, and interaction between proximity to type of food store selections with visits to determine association with %TWL over 24 months.

RESULTS : There were no significant differences for patients living within a 5- (p = 0.523) and 10-min (p = 0.580) walk in proximity to M/HD food selection stores and weight loss through 24 months. However, patients living in proximity to at least 1 LD selection store within a 5- (p = 0.027) and 1 or 2 LD stores within a 10-min (p = 0.015) walk had less weight loss through 24 months.

CONCLUSION : Overall, living in proximity to LD selection stores was a better predictor of postoperative weight loss over 24 months than living within proximity of M/HD selection stores.

Pratt Keeley J, Hanks Andrew S, Miller Harvey J, Swager LeeAnn C, Noria Sabrena, Brethauer Stacy, Needleman Bradley, Focht Brian C

2023-Feb-27

Bariatric surgery, Environment, Food, Walkability