Deep learning-based electroencephalic diagnosis of tinnitus symptom

Tinnitus is a neuropathological phenomenon caused by the recognition of external sound that does not actually exist. Existing diagnostic methods for tinnitus are rather subjective and complicated medical examination procedures. The present study aimed to diagnose tinnitus using deep learning analysis of electroencephalographic (EEG) signals while patients performed auditory cognitive tasks. We found that, during an active oddball task, patients with tinnitus could be identified with an area under the curve of 0.886 through a deep learning model (EEGNet) using EEG signals. Furthermore, using broadband (0.5 to 50 Hz) EEG signals, an analysis of the EEGNet convolutional kernel feature maps revealed that alpha activity might play a crucial role in identifying patients with tinnitus. A subsequent time-frequency analysis of the EEG signals indicated that the tinnitus group had significantly reduced pre-stimulus alpha activity compared with the healthy group. These differences were observed in both the active and passive oddball tasks. Only the target stimuli during the active oddball task yielded significantly higher evoked theta activity in the healthy group compared with the tinnitus group. Our findings suggest that task-relevant EEG features can be considered as a neural signature of tinnitus symptoms and support the feasibility of EEG-based deep-learning approach for the diagnosis of tinnitus.

Comparison between sitagliptin as add-on therapy to insulin and insulin dose-increase therapy in uncontrolled Korean type 2 diabetes: CSI study

AIM

Individuals requiring insulin therapy for type 2 diabetes often require escalation of their regimen to achieve glycaemic control. Optimal management strategies for uncontrolled type 2 diabetes would improve glycaemic control without hypoglycaemia and weight gain. This study compared the efficacy and tolerability of adding sitagliptin, an oral dipeptidyl peptidase-4 inhibitor, and an up to 20% increase in insulin dose in patients with uncontrolled type 2 diabetes on insulin therapy.

METHODS

We conducted a 24-week, randomized, active-competitor, parallel-group study in subjects with uncontrolled type 2 diabetes [haemoglobin A1c (HbA1c) = 7.5-11%] currently using insulin therapy. Subjects were randomly assigned to either the sitagliptin adding (100 mg daily, n = 70) or an insulin-increasing arm (≥ 10% at week 12 and ≥ 10% at week 24, n = 70) while continuing other medications.

RESULTS

Average baseline HbA1c was 9.2% in both groups. HbA1c decreased more at 24 weeks in the sitagliptin adding than the insulin-increasing arm (-0.6 ± 0.1% vs. -0.2 ± 0.1%, p < 0.01). Insulin was increased by 25% at 24 weeks in the insulin-increasing group. Hypoglycaemic events were less common and less severe in sitagliptin adding arm than insulin-increasing arm (7.0 vs. 14.3 events per patient-year, p < 0.05). Weight was stable in the sitagliptin adding subjects (68.6 ± 11.6 vs. 68.1 ± 11.4 kg) but increased in the insulin-increasing subjects (66.2 ± 10.6 vs. 67.4 ± 9.7 kg, p < 0.05). Other adverse events occurred at similar rates in both arms.

CONCLUSIONS

Compared to a 25% increase in insulin dose, adding sitagliptin to an insulin-based regimen was more effective at lowering HbA1c and associated with less hypoglycaemia and weight gain over 24 weeks.

CLINICAL TRIAL NUMBER

NCT01100125.