Banca de DEFESA: Lorena Silva Vieira

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : Lorena Silva Vieira
DATE: 30/08/2023
TIME: 14:00
LOCAL: Plataforma Teams
TITLE:

"Accuracy of artificial intelligence tools in tracking osteoporosis in dental imaging exams"

KEY WORDS:

"Osteoporosis; Artificial intelligence; dental radiograph; panoramic radiograph"

PAGES: 100
BIG AREA: Ciências da Saúde
AREA: Odontologia
SUMMARY:

"Introduction: Osteoporosis is a common bone metabolic disorder, which is characterized by progressive and silent concern of bone microarchitecture, and develops when there is a reduction in bone mineral density as well as bone quality. The diagnosis of osteoporosis is generally carried out using Bone Densitometry (Dual Energy X-Ray Absorptiometry (DXA), but because it is a costly test, it is still not accessible to most of the population. Therefore, auxiliary methods artificial intelligence (AI) has been proposed to help dentists in the screening of osteoporosis using dental imaging tests, such as radiographs and computed tomography. , which can be used to study the problem of osteoporosis.Objectives: to carry out a systematic review to evaluate the accuracy of artificial intelligence tools in tracking osteoporosis through dental imaging. Methodology: A search was carried out in PUBMED/Medline, Embase, LILACS, Web of Science, Scopus, Computers and Applied Sciences Complete (EBSCO), ACM Digital Library, Compendex, Google Scholar and Proquest. To search for studies, a combination of descriptors and free terms “Osteoporosis”, “artificial intelligence”, “dental radiograph”, “dental imaging” and “panoramic radiograph” were used. For selection and management of references, the EndNoteWeb® and Rayyan applications were used. Studies carried out with adult patients diagnosed with osteoporosis, and which applied artificial intelligence algorithms in the analysis of dental imaging exams, using DXA as a reference standard, were eligible. Articles that met the selection criteria were reviewed based on the QUADAS-2 guidelines and the certainty of the evidence was assessed using the GRADE approach. Results: A total of 799 studies were found, leaving 62 for the full reading, including 18 studies. AI has been widely applied for screening and diagnosis of osteoporosis. Of the analyzed studies, the Machine Learning (ML) tool was the most used, together with its Support Vector Machine (SVM) algorithm, which demonstrated better performance in the classification of osteoporosis using panoramic radiographs, when compared to the Naive Bayes and k- NN. Another tool used in the studies was Deep Learning (DL), which showed promising results through the VGG-16 and GoogLeNet algorithms. Conclusion: AI technologies prove to be excellent tools to identify patients at higher risk of developing osteoporosis, and thus can be used in clinical practice to help dentists in decision-making"

COMMITTEE MEMBERS:
Presidente - 3475960 - PAULO TADEU DE SOUZA FIGUEIREDO
Interna - 1889805 - CRISTINE MIRON STEFANI
Interna - 2315081 - NILCE SANTOS DE MELO
Externa à Instituição - NATHALIA FERRARE PINTO - OUTROS