Evaluation of alarm notification of artificial intelligence in automated analyzer detection of parasites.

Journal: Medicine
PMID:

Abstract

To evaluate the alarm notification of artificial intelligence in detecting parasites on the KU-F40 Fully Automatic Feces Analyzer and provide a reference for clinical diagnosis in parasite diseases. A total of 1030 fecal specimens from patients in our hospital from May to June 2023 were collected, and parasite detection studies were conducted using the KU-F40 automated feces analyzer (normal mode method, floating-sedimentation mode method), acid-ether sedimentation method, and direct smear microscopy method, respectively. The positive detection rate of parasites in the 1030 fecal specimens was 22.9% (236 cases), of which the KU-F40 normal mode method had a detection rate of 16.3% (168 cases), the acid-ether sedimentation method had a detection rate of 19.0% (196 cases), and the direct smear microscopy method had a detection rate of 13.1% (135 cases). The detection rates of the first 2 methods were higher than those of the direct smear microscopy method, and the difference was statistically significant (P < .05). The detection rate of the KU-F40 floating-sedimentation mode method was 11.9% (123 cases), which was lower than that of the direct smear microscopy, and the difference was not statistically significant (P > .05). The sensitivity of the KU-F40 normal mode method, acid-ether sedimentation method, direct smear microscopy method, and the KU-F40 floating-sedimentation mode method were 71.2%, 83.1%, 57.2%, and 52.1%, respectively, and the specificity was 94.7%, 100%, 100%, and 97.7%, respectively. The coincidence rates of the KU-F40 normal mode method was 90.78%, with Kappa values of 0.633. The positive detection rate of parasites using the KU-F40 normal mode method is higher than that using the direct smear microscopy method. It has high sensitivity and specificity and has advantages such as high automation and fast detection speed. It can replace the microscopy method for routine screening and has higher clinical application value in the diagnosis of intestinal parasitic diseases.

Authors

  • Zila Wang
    Key Laboratory of Clinical Laboratory Medicine of Guangxi Department of Education, Department of Clinical Laboratory, The First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi, China.
  • Lin Liao
    Department of Neurology, Nanfang Hospital, Southern Medical University, Guangzhou, China.
  • Ximei Huang
  • Jinguang Tang
  • Faquan Lin