OBJECTIVE: To compare the accuracy and reaction time of a new biopsy urease test, Pronto Dry (Medical Instruments Corporation, Solothurn, Switzerland) and the CLO test in the diagnosis of H. pylori infection.
METHODS: Consecutive patients presenting with dyspepsia to the endoscopy unit, University of Malaya Medical Centre were recruited for the study. Patients who were previously treated for H. pylori infection or who had received antibiotics, proton pump inhibitors or bismuth compounds in the preceding 4 weeks were excluded. H. pylori diagnosis was made based on the ultra rapid urease test and histological examination of gastric biopsies. Four antral and four corpus biopsies were taken for this purpose from all patients. A diagnosis of H. pylori infection was made when both the ultra rapid urease test and histology were positive in either the antral or corpus biopsies. A negative diagnosis of H. pylori was made when both tests from antral and corpus biopsies were all negative. Another four antral and four corpus biopsies (two each) were taken for the Pronto Dry and CLO tests. The Pronto Dry and CLO tests were stored and performed according to the manufacturer's instruction.
RESULTS: Two hundred and eight patients were recruited in the study. Eighty-six of the patients were males and 122 were females. The mean age was 46.3 years with a range of 15-82 years. The results for both the Pronto Dry and the CLO tests were completely concordant with sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of 98.1%, 100%, 100%, 98.1% and 99%, respectively. The Pronto Dry test showed a faster reaction time to positive compared with the CLO test, with 96.2% positive reaction by 30 min versus 70.8% and 100% positive reaction time by 55 min versus 83%. The colorimetric change was also more distinct with the Pronto Dry test compared with the CLO test.
CONCLUSIONS: Both the Pronto Dry and the CLO tests were highly accurate for the diagnosis of H. pylori infection. The Pronto Dry test showed a quicker positive reaction time and the positive colour change was more distinct.
RESULTS: Our models learned several syntactic, lexical, and n-gram linguistic biomarkers to distinguish the probable AD group from the healthy group. In contrast to the healthy group, we found that the probable AD patients had significantly less usage of syntactic components and significantly higher usage of lexical components in their language. Also, we observed a significant difference in the use of n-grams as the healthy group were able to identify and make sense of more objects in their n-grams than the probable AD group. As such, our best diagnostic model significantly distinguished the probable AD group from the healthy elderly group with a better Area Under the Receiving Operating Characteristics Curve (AUC) using the Support Vector Machines (SVM).
CONCLUSIONS: Experimental and statistical evaluations suggest that using ML algorithms for learning linguistic biomarkers from the verbal utterances of elderly individuals could help the clinical diagnosis of probable AD. We emphasise that the best ML model for predicting the disease group combines significant syntactic, lexical and top n-gram features. However, there is a need to train the diagnostic models on larger datasets, which could lead to a better AUC and clinical diagnosis of probable AD.