Diagnosis of diabetic kidney disease in whole slide images via AI-driven quantification of pathological indicators.

Diagnosis of diabetic kidney disease (DKD) mainly relies on screening the morphological variations and internal lesions of glomeruli from pathological kidney biopsy. The prominent pathological alterations of glomeruli for DKD include glomerular hypertrophy and nodular mesangial sclerosis. However, the qualitative judgment of these alterations is inaccurate and inconstant due to the intra- and inter-subject variability of pathologists. It is necessary to design artificial intelligence (AI) methods for accurate quantification of these pathological alterations and outcome prediction of DKD. In this work, we present an AI-driven framework to quantify the volume of glomeruli and degree of nodular mesangial sclerosis, respectively, based on an instance segmentation module and a novel weakly supervised Macro-Micro Aggregation (MMA) module. Subsequently, we construct classic machine learning models to predict the degree of DKD based on three selected pathological indicators via factor analysis. These corresponding modules are trained and tested on a total of 281 whole slide images (WSIs) digitized from two hospitals with different scanners. Our designed AI framework achieved inspiring results with 0.926 mIoU for glomerulus segmentation, and 0.899 F1 score for glomerulus classification in the external testing dataset. Meantime, the visualized results of the MMA module could reflect the location of the lesions. The performance of predicting disease achieved the F1 score of 0.917, which further proved the effectiveness of our AI-driven quantification of pathological indicators. Additionally, the interpretation of the machine learning model with the SHAP method showed similar accordance with the development of DKD in pathology. In conclusion, the proposed auxiliary diagnostic technologies have the feasibility for quantitative analysis of glomerular pathological tissues and alterations in DKD. Pathological quantitative indicators will also make it more convenient to provide doctors with assistance in clinical practice.

A Multifactorial Risk Score System for the Prediction of Diabetic Kidney Disease in Patients with Type 2 Diabetes Mellitus.

Purpose: In-depth investigations of risk factors for the identification of diabetic kidney disease (DKD) in type 2 diabetes mellitus (T2DM) are rare. We aimed to investigate the risk factors for developing DKD from multiple types of clinical data and conduct a comprehensive risk assessment for individuals with diabetes. Methods: We carried out a case-control study, enrolling 958 patients to identify the risk factors for developing DKD in T2DM patients from a database established from inpatient electronic medical records. Multivariable logistic regression was applied to develop a prediction model and the performance of the model was evaluated using the area under the curve (AUC) and calibration curve. A multifactorial risk score system was established according to the Framingham Study risk score. Results: DKD accounted for 34.03% of eligible patients in total. Twelve risk factors were selected in the final prediction model, including age, duration of diabetes, duration of hypertension, fasting blood glucose, fasting C-peptide, insulin use, systolic blood pressure, low-density lipoprotein, γ-glutamyl transpeptidase, platelet, uric acid, and thyroid stimulating hormone; and one protective factor, serum albumin. The prediction model showed an AUC of 0.862 (95% Confidence Interval (CI) 0.834-0.890) with an accuracy of 81.5% in the derivation dataset and an AUC of 0.876 (95% CI 0.825-0.928) in the validation dataset. The calibration curves were excellent and the estimated probability of DKD was more than 80% when the cumulative score for risk factors reached 17 points. Conclusion: Newly recognized risk factors were applied to assess the development of DKD in T2DM patients and the established risk score system was a reliable and feasible tool for assisting clinicians to identify patients at high risk of DKD.

Analysis of clinical predictors of kidney diseases in type 2 diabetes patients based on machine learning.

BACKGROUND: The heterogeneity of Type 2 Diabetes Mellitus (T2DM) complicated with renal diseases has not been fully understood in clinical practice. The purpose of the study was to propose potential predictive factors to identify diabetic kidney disease (DKD), nondiabetic kidney disease (NDKD), and DKD superimposed on NDKD (DKD + NDKD) in T2DM patients noninvasively and accurately. METHODS: Two hundred forty-one eligible patients confirmed by renal biopsy were enrolled in this retrospective, analytical study. The features composed of clinical and biochemical data prior to renal biopsy were extracted from patients' electronic medical records. Machine learning algorithms were used to distinguish among different kidney diseases pairwise. Feature variables selected in the developed model were evaluated. RESULTS: Logistic regression model achieved an accuracy of 0.8306 ± 0.0057 for DKD and NDKD classification. Hematocrit, diabetic retinopathy (DR), hematuria, platelet distribution width and history of hypertension were identified as important risk factors. Then SVM model allowed us to differentiate NDKD from DKD + NDKD with accuracy 0.8686 ± 0.052 where hematuria, diabetes duration, international normalized ratio (INR), D-Dimer, high-density lipoprotein cholesterol were the top risk factors. Finally, the logistic regression model indicated that DD-dimer, hematuria, INR, systolic pressure, DR were likely to be predictive factors to identify DKD with DKD + NDKD. CONCLUSION: Predictive factors were successfully identified among different renal diseases in type 2 diabetes patients via machine learning methods. More attention should be paid on the coagulation factors in the DKD + NDKD patients, which might indicate a hypercoagulable state and an increased risk of thrombosis.

Morula transfer achieves better clinical outcomes than post-thawed cleavage embryos after overnight culture in frozen embryo transfer (FET) cycles.

PURPOSE: This study aimed to investigate the clinical outcomes of morula stage transfer derived from post-thawed cleavage embryos undergoing overnight culture in frozen embryo transfer (FET) cycles. METHODS: We performed a retrospective study that included 392 FET cycles with 784 thawed embryos undergoing overnight culture between January 2014 and December 2018. Embryos were divided into three groups in terms of their status: 8-16 cells without morula (group I), one morula (group II), and two morulae (group III). The clinical outcomes of these cycles were then compared between the three groups. Logistic regression analysis was performed to control for confounders. RESULTS: Group III was associated with a significantly higher clinical pregnancy rate (odds ratio [OR] 2.35; 95% confidence interval [CI] 1.29-4.27; P = 0.005), implantation rate (OR 3.00; CI 1.75-5.16; P < 0.001), multiple pregnancy rate (OR 4.91; CI 2.11-11.40; P < 0.001), and live birth rate (OR 1.96; CI 1.10-3.49; P = 0.022) than group I. Group II had a higher live birth rate than group I after adjustment (OR 1.70; CI 1.04-2.79; P = 0.035). There was no difference in the rate of premature delivery when compared across the three groups after adjustment. CONCLUSION: The transfer of morula stage embryos following the overnight culture of post-thawed cleavage embryos led to an improvement in the clinical outcomes of FET cycles. It is important to reduce the number of morula embryos transferred in order to achieve a singleton pregnancy.