OBJECTIVES: To construct vocal recognition classification models using 6 machine learning algorithms and vocal emotional characteristics of individuals with subthreshold depression to facilitate early identification of subthreshold depression. METHODS: We collected voice data from both normal individuals and participants with subthreshold depression by asking them to read specifically chosen words and texts. From each voice sample, 384-dimensional vocal emotional feature variables were extracted, including energy feature, Meir frequency cepstrum coefficient, zero cross rate feature, sound probability feature, fundamental frequency feature, difference feature. The Recursive Feature Elimination (RFE) method was employed to select voice feature variables. Classification models were then built using the machine learning algorithms Adaptive Boosting (AdaBoost), Random Forest (RF), Linear Discriminant Analysis (LDA), Logistic Regression (LR), Lasso Regression (LRLasso), and Support Vector Machine (SVM), and the performance of these models was evaluated. To assess generalization capability of the models, we used real-world speech data to evaluate the best speech recognition classification model. RESULTS: The AdaBoost, RF, and LDA models achieved high prediction accuracies of 100%, 100%, and 93.3% on word-reading speech test set, respectively. In the text-reading speech test set, the accuracies of the AdaBoost, RF, and LDA models were 90%, 80%, and 90%, respectively, while the accuracies of the other 3 models were all below 80%. On real-world word-reading and text-reading speech data, the classification models using AdaBoost and Random Forest still achieved high predictive accuracies (91.7% and 80.6% for AdaBoost and 86.1% and 77.8% for Random, respectively). CONCLUSIONS Analyzing vocal emotional characteristics allows effective identification of individuals with subthreshold depression. The AdaBoost and RF models show excellent performance for classifying subthreshold depression individuals, and may thus potentially offer valuable assistance in the clinical and research settings.
Humans ; Machine Learning ; Emotions ; Depression/diagnosis* ; Algorithms ; Voice ; Support Vector Machine ; Male ; Female

Objectives To investigate the efficacy and safety of quadruple regimen containing doxycy-cline,amoxicillin and clavulanate potassium in the treatment of Helicobacter pylori(Hp)after the failed initial eradication. Methods A total of 136 patients who had a failed course of initial eradication therapy were randomly divided into four groups. The 46 patients in group A were given rabeprazole,colloidal bismuth tartrate,doxycy-cline,amoxicillin and clavulanate potassium for 14 days). The 45 cases in group B were treated with rabeprazole, colloidal bismuth tartrate,doxycycline,amoxicillin and clavulanate potassium for 10 days. The 45 cases in control group received rabeprazole,colloidal bismuth tartrate,levofloxacin,amoxicillin for 14 days. After 4 weeks of treat-ment,the 13C-urea breath test was performed. The Hp eradication rates,adverse reactions,treatment compliance and cost-effectiveness ratio of the three groups were evaluated. Results A total of 130 patients completed the study. The intention to treat(ITT)and Per-protocol(PP)analyses revealed that the eradication rates of the three groups were 71.74% and 76.74%(group A),71.11% and 72.72%(group B),51.11% and 53.49%(control group),respectively. The ITT and PP of group were significantly higher than those of control group(P < 0.05). But there was no significant difference between group B and control group(P > 0.05). The rates of adverse reac-tions in each group were not statistically significant(P > 0.05),and the patient compliance of three groups was more than 80%. The cost and cost-effectiveness ratio showed that group B < control group < group A. Conclu-sions The quadruple therapy containing doxycycline,amoxicillin and clavulanate potassium as a rescue regimen given for 14 days is safe,effective and with high patient compliance,which is of high clinical significance.

Objective To explore the value of miniprobe sonography (MPS) in differentiating Crohn??s disease (CD) and ulcerative colitis (UC). Methods 46 patients with active inflammatory bowel disease (IBD) were included into the patient group (16 patients with CD and 30 cases with UC). 20 healthy cases ( underwent physical examination in outpatient department) were recruited as the control group. All cases underwent MPS and high sensitive C-reactive protein (Hs-CRP) test.The MPS measurement index include thetotalwallthickness(TWT), mucosal thickness (MT), submucosal thickness (SMT), muscularispropria thickness (MPT), and the number of enlarged colic and paracolic lymph nodes. Results TheTWT, M, SM, MP(mm)and Hs-CRP(mg/L)in CD, UC and control group are 5.84 ± 1.42, 1.48 ± 0.23, 1.88 ± 0.28, 2.31 ± 0.85, 40.58 ± 19.33, 4.74 ± 1.01, 1.79 ± 0.35, 1.41 ± 0.25, 1.32 ± 0.34, 22.41 ± 15.25,2.86 ± 0.23, 0.97 ± 0.13, 1.06 ± 0.11, 0.64 ± 0.0 and 1.70 ± 0.65. TWT, MT, SMT, MPT and Hs-CRP in UC group has significant different with those in controp group(P<0.05). The mean value of TWT, MT, SMT, MPT and Hs-CRP in CD group are higher than those in UC group. M in UC group is higher than that in CE group. The difference is significant(P<0.05). 5 cases in CD and 2 cases with UC had enlarged colic or paracolic lymph nodes. Conclusions The MPS can distinctly observe the changesof different colonic tissue layers and the surrounding tissue structures in IBD patients,which helps for distinguishing active CD from UC.