Artificial intelligence-enhanced electrocardiogram analysis for identifying cardiac autonomic neuropathy in patients with diabetes.

AIM: To develop and employ machine learning (ML) algorithms to analyse electrocardiograms (ECGs) for the diagnosis of cardiac autonomic neuropathy (CAN). MATERIALS AND METHODS: We used motif and discord extraction techniques, alongside long short-term memory networks, to analyse 12-lead, 10-s ECG tracings to detect CAN in patients with diabetes. The performance of these methods with the support vector machine classification model was evaluated using 10-fold cross validation with the following metrics: accuracy, precision, recall, F1 score, and area under the receiver-operating characteristic curve (AUC). RESULTS: Among 205 patients (mean age 54 ± 17 years, 54% female), 100 were diagnosed with CAN, including 38 with definite or severe CAN (dsCAN) and 62 with early CAN (eCAN). The best model performance for dsCAN classification was achieved using both motifs and discords, with an accuracy of 0.92, an F1 score of 0.92, a recall at 0.94, a precision of 0.91, and an excellent AUC of 0.93 (95% confidence interval [CI] 0.91-0.94). For the detection of any stage of CAN, the approach combining motifs and discords yielded the best results, with an accuracy of 0.65, F1 score of 0.68, a recall of 0.75, a precision of 0.68, and an AUC of 0.68 (95% CI 0.54-0.81). CONCLUSION: Our study highlights the potential of using ML techniques, particularly motifs and discords, to effectively detect dsCAN in patients with diabetes. This approach could be applied in large-scale screening of CAN, particularly to identify definite/severe CAN where cardiovascular risk factor modification may be initiated.

Be cautious during the interpretation of arterial blood gas analysis performed outside the intensive care unit.

INTRODUCTION: Reliable results of an arterial blood gas (ABG) analysis are crucial for the implementation of appropriate diagnostics and therapy. We aimed to investigate the differences (Δ) between ABG parameters obtained from point-of-care testing (POCT) and central laboratory (CL) measurements, taking into account the turnaround time (TAT). MATERIALS AND METHODS: A number of 208 paired samples were collected from 54 intensive care unit (ICU) patients. Analyses were performed using Siemens RAPIDPoint 500 Blood Gas System on the samples just after blood retrieval at the ICU and after delivery to the CL. RESULTS: The median TAT was 56 minutes (IQR 39-74). Differences were found for all ABG parameters. Median Δs for acid-base balance ere: ΔpH=0.006 (IQR -0.0070-0.0195), ΔBEef=-0.9 (IQR -2.0-0.4) and HCO3-act=-1.05 (IQR -2.25-0.35). For ventilatory parameters they were: ΔpO2=-8.3 mmHg (IQR -20.9-0.8) and ΔpCO2=-2.2 mmHg (IQR -4.2--0.4). For electrolytes balance the differences were: ΔNa+=1.55 mM/L (IQR 0.10-2.85), ΔK+=-0.120 mM/L (IQR -0.295-0.135) and ΔCl-=1.0 mM/L (IQR -1.0-3.0). Although the Δs might have caused misdiagnosis in 51 samples, Bland-Altman analysis revealed that only for pO2 the difference was of clinical significance (mean: -10.1 mmHg, ±1.96SD -58.5; +38.3). There was an important correlation between TAT and ΔpH (R=0.45, p<0.01) with the safest time delay for proper assessment being less than 39 minutes. CONCLUSIONS: Differences between POCT and CL results in ABG analysis may be clinically important and cause misdiagnosis, especially for pO2. POCT should be advised for ABG analysis due to the impact of TAT, which seems to be the most important for the analysis of pH.