Qualitative Research in Dentistry Using Natural Language Understanding (NLU)
Objectives: Background Natural Language Understanding (NLU) is a subfield of computer science concerned with the usage of computational techniques to learn, understand, and produce human language content, and it could provide an objective quantitative assessment of interviews in qualitative research. This approach could be useful for qualitative research in oral health research. The objective of this study was to assess the use of NLU for research on the impact of oral health conditions (i.e. hypodontia) on patients’ quality of life. Methods: Methods: The study design was qualitative research using NLU for thematic analysis and compared to the data were transcribed and analysed using Nvivo soft wear. Semi-structured interviews were conducted on 10 participants aged 16–25 years (five males and five females) suffering from hypodontia, and the interviewees’ transcripts were recorded. The Transcripts were then analysed using IBM Watson Natural Language Understanding Text Analysis. Automated analysis for the sentiment produced weighted valence (extent of positive, negative or neutral emotive implications) for specific entities identified in the transcripts. The results of the NLU analysis was compared with a traditional qualitative assessment of the transcripts performed by an independent group of researchers blinded to the NLU analysis. Results: Result: NLU analysis of the interview transcripts showed a predominantly negative sentiment towards the hypodontia and its management; 95.3% of the entities identified by the NLU in the transcript were related to a negative sentiment while only 4.7% had a positive sentiment. The most frequent negative entities were related to treatment waiting time (69%), aesthetics (28%) and lack of communication (3%). These results were similar to those obtained by traditional qualitative research analysis which also showed participants dissatisfaction with treatment waiting time and lack of communication concerns with appearance, and psychosocial well-being. Conclusions: Conclusion: NLU could be a useful tool for qualitative research in dentistry.
Division: Meeting:2022 IADR/APR General Session (Virtual) Location: Year: 2022 Final Presentation ID:1636 Abstract Category|Abstract Category(s):e-Oral Health Network
Authors
Anweigi, Lamyia
( Qatar University
, Doha
, Dental college Qatar university
, Qatar
)
Ahmeda, Mohamed
( Sherborne Qatar: International British Schools
, Doha
, Qatar
)
Awad, Jomana
( Qatar University
, Doha
, Dental college Qatar university
, Qatar
)
Ben Naceur, Iheb
( Qatar University
, Doha
, Dental college Qatar university
, Qatar
)
Barhom, Noha
( Qatar University
, Doha
, Qatar
)
Tamimi, Faleh
( Qatar University
, Doha
, Dental college Qatar university
, Qatar
)
Financial Interest Disclosure: none
SESSION INFORMATION
Interactive Talk Session
e-Oral Health Network I
Saturday,
06/25/2022
, 02:00PM - 03:30PM