Related Experiment Video
Updated: Aug 3, 2026
05:56
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
Published on: April 14, 2023
[Exemplary classification of otolaryngologic data using the Kohonen artificial neural network]
Dariusz Kaczmarczyk1, Marcin Durko, Andrzej Kruk
1Uniwersytet Medyczny w Lodzi, Zakład Cytofizjologii, Histologii i Embriologii. kaczmard@csk.am.lodz.pl
Polski Merkuriusz Lekarski : Organ Polskiego Towarzystwa Lekarskiego
|December 20, 2005
Summary
Kohonen artificial neural networks effectively classify otosclerosis patient data, identifying three distinct groups based on hearing function. This classification aids in understanding preoperative and postoperative outcomes for otosclerosis patients.
Area of Science:
- Otolaryngology
- Artificial Intelligence
- Biomedical Engineering
Background:
- Otosclerosis diagnosis and treatment can be complex.
- Objective classification methods are needed to stratify patients.
- Kohonen artificial neural networks offer a potential solution for data analysis.
Purpose of the Study:
- To classify otolaryngological data from otosclerosis patients using Kohonen artificial neural networks.
- To evaluate the effectiveness of this classification in grouping patients.
- To identify distinct patient clusters based on audiological and surgical outcomes.
Main Methods:
- Utilized Kohonen artificial neural network for data classification.
- Analyzed data from 220 otosclerosis patients (184 female, 36 male, aged 16-68).
- Evaluated classification effectiveness based on preoperative and postoperative air-bone gap, and other clinical variables.
Main Results:
- The neural network successfully distinguished 3 distinct patient clusters.
- Clusters showed significant differences in preoperative air and bone conduction.
- Postoperative air-bone gap values varied considerably between the identified clusters.
Conclusions:
- Kohonen artificial neural networks provide an effective method for classifying otosclerosis patient data.
- The identified clusters correlate with known differences in otosclerosis patient characteristics and outcomes.
- This AI-driven approach can enhance understanding and potentially guide treatment strategies for otosclerosis.
