AI Detection Model of Extra Root Canals in Mandibular Premolars Using CBCT Scans
Diagnostic Accuracy of a Deep Learning Model (Artificial Intelligence) for Detecting Extra Root Canals in Mandibular Premolars on CBCT Images: Diagnostic Accuracy Study.
Successful endodontic treatment depends on the complete identification and management of the entire root canal system. Missed root canals are a major cause of endodontic failure, particularly in mandibular premolars, which exhibit considerable anatomical variability and may contain additional root canals that are difficult to detect using conventional diagnostic methods.
Cone Beam Computed Tomography (CBCT) provides three-dimensional visualization of root canal anatomy and has significantly improved the detection of anatomical variations. However, interpretation of CBCT images remains dependent on the experience and expertise of the clinician, leading to potential observer variability and missed diagnoses.
Recent advances in artificial intelligence (AI), particularly deep learning models based on convolutional neural networks, have shown promising results in dental image analysis and diagnostic support. AI-assisted diagnostic systems may improve the accuracy, consistency, and efficiency of CBCT interpretation by automatically identifying complex anatomical structures.
The aim of this retrospective diagnostic accuracy study is to evaluate the performance of a newly developed deep learning model for the detection of extra root canals in mandibular premolars using CBCT images. The diagnostic accuracy of the AI model will be assessed by comparing its findings with the assessments of experienced oral and maxillofacial radiologists, which will serve as the reference standard.
A total of 272 CBCT scans of mandibular premolars from Egyptian patients will be included according to predefined eligibility criteria. Diagnostic performance will be evaluated using measures including sensitivity, specificity, positive predictive value, and negative predictive value.
The findings of this study may provide evidence regarding the clinical applicability of AI-assisted diagnostic tools in endodontics and contribute to improved detection of complex root canal anatomy, reduced incidence of missed canals, and enhanced treatment outcomes.
Studieoversikt
Status
Forhold
Intervensjon / Behandling
Detaljert beskrivelse
The goal of this observational study is to evaluate whether a deep learning artificial intelligence (AI) model can accurately detect extra root canals in mandibular premolars using Cone Beam Computed Tomography (CBCT) images in Egyptian patients. The main questions it aims to answer are:
- Can the AI model accurately detect extra root canals in mandibular premolars on CBCT scans?
- Is the diagnostic accuracy of the AI model comparable to that of experienced oral and maxillofacial radiologists? Researchers will compare the results generated by the AI model with the assessments of experienced radiologists, which will serve as the reference standard.
Participants will:
- Provide previously acquired CBCT scans that meet the study eligibility criteria.
- Have their CBCT images analyzed by the AI model.
- Have their CBCT images independently evaluated by experienced radiologists for comparison with the AI findings.
The study findings may help determine the potential role of AI-assisted diagnostic tools in improving the detection of complex root canal anatomy and supporting endodontic diagnosis
Studietype
Registrering (Antatt)
Fase
- Ikke aktuelt
Kontakter og plasseringer
Studiekontakt
- Navn: Ayah Tarek, PHD candidate
- Telefonnummer: 20201221902479
- E-post: ayahtarek94@gmail.com
Deltakelseskriterier
Kvalifikasjonskriterier
Alder som er kvalifisert for studier
- Voksen
- Eldre voksen
Tar imot friske frivillige
Beskrivelse
Inclusion Criteria:
- CBCT scans of mandibular molars of Egyptian patients aging from 18 to 65 years old
- Small Field of view (FOV) including maximum a quadrant
- Voxel size not larger than 2mm
- Mandibular premolars showing complete root formation
- Carious or non-carious teeth
- Absence of artifacts.
Exclusion Criteria:
- Mandibular first and second premolars with developmental anomalies, external or internal root resorption, root canal calcification, previous root canal treatment, post restorations, and/or root caries
- CBCT images of sub-optimal quality or artifacts/high scatter interfering with proper assessment
Studieplan
Hvordan er studiet utformet?
Designdetaljer
- Primært formål: Diagnostisk
- Tildeling: Randomisert
- Intervensjonsmodell: Parallell tildeling
- Masking: Ingen (Open Label)
Våpen og intervensjoner
Deltakergruppe / Arm |
Intervensjon / Behandling |
|---|---|
|
Eksperimentell: Mandibular premolars with single canals
|
It is a study to detect the diagnostic accuracy of AI model to detect extra canals in mandibular premolars
|
|
Eksperimentell: Mandibular premolars with more than one canal
|
It is a study to detect the diagnostic accuracy of AI model to detect extra canals in mandibular premolars
|
Hva måler studien?
Primære resultatmål
Resultatmål |
Tiltaksbeskrivelse |
Tidsramme |
|---|---|---|
|
Diagnostic Accuracy of the Deep Learning Model for Detection of Extra Root Canals in Mandibular Premolars
Tidsramme: During the procedure
|
Diagnostic accuracy of the AI model will be determined by comparison with expert radiologist assessment.
|
During the procedure
|
Sekundære resultatmål
Resultatmål |
Tidsramme |
|---|---|
|
Sensitivity of the AI Model Specificity of the AI Model Positive Predictive Value (PPV) Negative Predictive Value (NPV)
Tidsramme: During the procedure
|
During the procedure
|
Samarbeidspartnere og etterforskere
Sponsor
Publikasjoner og nyttige lenker
Studierekorddatoer
Studer hoveddatoer
Studiestart (Antatt)
Primær fullføring (Antatt)
Studiet fullført (Antatt)
Datoer for studieregistrering
Først innsendt
Først innsendt som oppfylte QC-kriteriene
Først lagt ut (Faktiske)
Oppdateringer av studieposter
Sist oppdatering lagt ut (Faktiske)
Siste oppdatering sendt inn som oppfylte QC-kriteriene
Sist bekreftet
Mer informasjon
Begreper knyttet til denne studien
Nøkkelord
Andre studie-ID-numre
- 7.1.1
Plan for individuelle deltakerdata (IPD)
Planlegger du å dele individuelle deltakerdata (IPD)?
Legemiddel- og utstyrsinformasjon, studiedokumenter
Studerer et amerikansk FDA-regulert medikamentprodukt
Studerer et amerikansk FDA-regulert enhetsprodukt
Denne informasjonen ble hentet direkte fra nettstedet clinicaltrials.gov uten noen endringer. Hvis du har noen forespørsler om å endre, fjerne eller oppdatere studiedetaljene dine, vennligst kontakt register@clinicaltrials.gov. Så snart en endring er implementert på clinicaltrials.gov, vil denne også bli oppdatert automatisk på nettstedet vårt. .