Prediction of Diabetic Sensorimotor Polyneuropathy Using Machine Learning Techniques

Identification of blood metabolic biomarkers associated with diabetic distal symmetric sensorimotor polyneuropathy in patients with type 2 diabetes mellitus

BACKGROUND

Distal symmetric sensorimotor polyneuropathy (DSPN) is a common neurologic complication of type 2 diabetes mellitus (T2DM), but the underlying mechanisms and changes in serum metabolites remain largely undefined. This study aimed to characterize the plasma metabolite profiles of participants with T2DM using targeted metabolomics analysis and identify potential biomarkers for DSPN.

METHODS

A combined liquid chromatography MS/MS and direct flow injection were used to quantify plasma metabolite obtained from 63 participants with T2DM, 81 with DSPN, and 33 nondiabetic control participants. A total of 130 metabolites, including amino acids, biogenic amines, sphingomyelins (SM), phosphatidylcholines, carnitines, and hexose, were analyzed.

RESULTS

A total of 16 plasma metabolites and 3 cholesterol-related laboratory parameters were found to have variable importance in the projection score >1.0 and false discovery rate <5.0% between control, T2DM, and DSPN. Among these variables, five serum metabolites, including phenylalanine (AUC = 0.653), alanine (AUC = 0.630), lysine (AUC = 0.622) tryptophan (AUC = 0.620), and SM C16:0 (AUC = 0.630), are potential biomarkers (all p < .05) in distinguishing T2DM with DSPN from those without (AUC = 0.720).

CONCLUSIONS

In this cross-sectional study, derangement of several metabolites in the plasma was observed in T2DM with and without DSPN, and these metabolites may be potential biomarkers for predicting DSPN. Longitudinal studies are warranted.

Development and validation of risk prediction model for diabetic neuropathy among diabetes mellitus patients at selected referral hospitals, in Amhara regional state Northwest Ethiopia, 2005-2021

BACKGROUND

Diabetic neuropathy is the most common complication in both Type-1 and Type-2 DM patients with more than one half of all patients developing nerve dysfunction in their lifetime. Although, risk prediction model was developed for diabetic neuropathy in developed countries, It is not applicable in clinical practice, due to poor data, methodological problems, inappropriately analyzed and reported. To date, no risk prediction model developed for diabetic neuropathy among DM in Ethiopia, Therefore, this study aimed prediction the risk of diabetic neuropathy among DM patients, used for guiding in clinical decision making for clinicians.

OBJECTIVE

Development and validation of risk prediction model for diabetic neuropathy among diabetes mellitus patients at selected referral hospitals, in Amhara regional state Northwest Ethiopia, 2005-2021.

METHODS

A retrospective follow up study was conducted with a total of 808 DM patients were enrolled from January 1,2005 to December 30,2021 at two selected referral hospitals in Amhara regional state. Multi-stage sampling techniques were used and the data was collected by checklist from medical records by Kobo collect and exported to STATA version-17 for analysis. Lasso method were used to select predictors and entered to multivariable logistic regression with P-value<0.05 was used for nomogram development. Model performance was assessed by AUC and calibration plot. Internal validation was done through bootstrapping method and decision curve analysis was performed to evaluate net benefit of model.

RESULTS

The incidence proportion of diabetic neuropathy among DM patients was 21.29% (95% CI; 18.59, 24.25). In multivariable logistic regression glycemic control, other comorbidities, physical activity, hypertension, alcohol drinking, type of treatment, white blood cells and red blood cells count were statistically significant. Nomogram was developed, has discriminating power AUC; 73.2% (95% CI; 69.0%, 77.3%) and calibration test (P-value = 0.45). It was internally validated by bootstrapping method with discrimination performance 71.7 (95% CI; 67.2%, 75.9%). It had less optimism coefficient (0.015). To make nomogram accessible, mobile based tool were developed. In machine learning, classification and regression tree has discriminating performance of 70.2% (95% CI; 65.8%, 74.6%). The model had high net benefit at different threshold probabilities in both nomogram and classification and regression tree.

CONCLUSION

The developed nomogram and decision tree, has good level of accuracy and well calibration, easily individualized prediction of diabetic neuropathy. Both models had added net benefit in clinical practice and to be clinically applicable mobile based tool were developed.

Recent applications of machine learning and deep learning models in the prediction, diagnosis, and management of diabetes: a comprehensive review

Diabetes as a metabolic illness can be characterized by increased amounts of blood glucose. This abnormal increase can lead to critical detriment to the other organs such as the kidneys, eyes, heart, nerves, and blood vessels. Therefore, its prediction, prognosis, and management are essential to prevent harmful effects and also recommend more useful treatments. For these goals, machine learning algorithms have found considerable attention and have been developed successfully. This review surveys the recently proposed machine learning (ML) and deep learning (DL) models for the objectives mentioned earlier. The reported results disclose that the ML and DL algorithms are promising approaches for controlling blood glucose and diabetes. However, they should be improved and employed in large datasets to affirm their applicability.

Clinical Research on Type 2 Diabetes: A Promising and Multifaceted Landscape

Type 2 diabetes constitutes an imposing epidemiological, economic, and scientific global challenge [...]

Addressing Chemotherapy-Induced Peripheral Neuropathy Using Multi-Frequency Vibrometry and Patient-Reported Outcomes

(1) The study evaluated correlations between multi-frequency vibrometry (MF-V) and the measure of chemotherapy-induced peripheral neuropathy developed by the European Organization for the Research and Treatment of Cancer (CIPN18). (2) Patients with cancer scheduled to undergo treatment with capecitabine and oxaliplatin (CAPOX) or carboplatin and paclitaxel (Carbo-Tax) were recruited in a prospective, observational study with MF-V and the CIPN18 from baseline to one year after end of treatment. (3) The study recruited 31 evaluable patients. All MF-V measurements correlated significantly with the CIPN18 scores (r = 0.25−0.48, p > 0.003), with a low frequency (32 Hz) from metatarsals showing the best correlation coefficients (0.059 Z-score per CIPN18 point change, r = 0.48, CI-95 = [0.32; 0.60], p > 0.0001). The largest change in MF-V scores from baseline was seen in low-frequency VPTs taken from metatarsals at 8 Hz three months after end of treatment (from −0.26, CI-95 [−0.85, 0.38] to 1.15, CI-95 [0.53, 1.84]) for patients treated with oxaliplatin and at 32 Hz one year after end of treatment (from 0.09, CI-95 [−0.56, 0.77] to 0.88, CI-95 [0.34, 1.47]) for patients treated with paclitaxel. (4) Low-frequency vibration perception thresholds (8 and 32 Hz) correlated better with CIPN18 scores than high-frequency ones (128 and 250 Hz). If validated, this finding will advance CIPN pathophysiological understanding and inform the development of assessment methods.