Name: | Description: | Size: | Format: | |
---|---|---|---|---|
Projeto de pós-graduação_39473 | 1.01 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
A inteligência artificial tem tido um crescimento exponencial nestes últimos anos e começa a ser usada em várias áreas da medicina dentária, nomeadamente na endodontia. Na área da endodontia tem tido uma grande evolução no sentido de permitir auxiliar o diagnóstico das infeções pulpares, baseando-se em análises radiográficas ou em scanners dentários. Os programas de inteligência artificial, de reconhecimento radiográfico, usam volumosos conjuntos de dados para estabelecer os seus diagnósticos e permitir previsões de evolução das doenças periapicais. As bases de dados usadas pelos softwares de inteligência artificial são continuamente melhoradas e expandidas à medida que estes são usados, tornando-se cada vez mais precisos. Outro aspeto relevante é a possibilidade de, através da inteligência artificial, termos a possibilidade de otimização dos tratamentos graças a sistemas capazes de criar planos de tratamentos personalizados e adaptados a cada paciente em função das suas próprias características anatómicas. A inteligência artificial ainda está no início da sua era, mas já permite detetar lesões periapicais, fragmentos de limas intracanalares, perda óssea, malformações dentárias, tratamentos já realizados, lesões cariosas, estruturas anatómicas, etc. Alguns softwares já podem prever a evolução de doenças dentárias e as eventuais necessidades de tratamentos. O objetivo desta revisão narrativa é avaliar a pertinência e o impacto do potencial uso da inteligência artificial na endodontia, identificando os objetivos, as várias perspetivas e os limites da sua utilização permitindo a realização de um correto diagnóstico e plano de tratamento e, assim, uma prática da endodontia mais eficaz. A pesquisa bibliográfica foi realizada através de motores de busca como a PubMed, Science Direct/Springer Link, B-on, e resumos de atas de conferências e congressos. Durante a pesquisa, foram usados vários termos MeSH (Medical Subject Headings) que correspondam às seguintes palavras-chave: Artificial intelligence/IA/AI, endodontics, machine learning, Convolutional Neural Network/CNN, apical lesion, root canal treatment, IA-guided endodontics, pulpectomy, em inglês; combinadas entre si com os operadores booleanos “OR” e “AND”. Os critérios de inclusão foram: artigos publicados nos últimos cinco anos (2019–2024), artigos que abordassem as diferentes perspetivas da utilização da inteligência artificial na endodontia, artigos que analisassem a relação entre IA e endodontia, artigos publicados após o início do uso da IA em endodontia. Foram excluídos os artigos que não cumprissem com os critérios de inclusão definidos. Os critérios de exclusão foram: artigos duplicados, artigos que necessitavam de custos adicionais para acesso sem texto integral disponível, e estudos não claros e explicitados. Pretendeu-se com esta revisão de literatura sintetizar as descobertas feitas em relação à inteligência artificial na prática endodôntica nestes últimos 5 anos, criando assim uma fonte de informação preciosa para os profissionais de saúde que procuram aplicações e perspetivas da inteligência artificial na endodontia.
Artificial intelligence has experienced exponential growth in recent years and is beginning to be used in various areas of dentistry, particularly in endodontics. In the field of endodontics, it has significantly evolved to assist in the diagnosis of pulp infections, based on radiographic analyses or dental scanners. Radiographic recognition AI programs use large datasets to establish their diagnoses and predict the progression of periapical diseases. The databases used by AI software are continuously improved and expanded as they are used, becoming increasingly precise. Another relevant aspect is the possibility, through artificial intelligence, of optimizing treatments thanks to systems capable of creating personalized treatment plans adapted to each patient based on their own anatomical characteristics. Artificial intelligence is still in its early stages but already enables the detection of periapical lesions, intracanal file fragments, bone loss, dental malformations, previously performed treatments, carious lesions, anatomical structures, etc. Some software can already predict the progression of dental diseases and the potential need for treatments. The objective of this narrative review is to evaluate the relevance and impact of the potential use of artificial intelligence in endodontics by identifying the objectives, various perspectives, and limits of its use, enabling accurate diagnosis and treatment planning and, thus, more effective endodontic practice. The bibliographic research will be conducted using search engines such as PubMed, Science Direct/Springer Link, B-on, and conference and congress abstract summaries. During the research, various MeSH (Medical Subject Headings) terms corresponding to the following keywords were used: Artificial intelligence/IA/AI, endodontics, machine learning, Convolutional Neural Network/CNN, apical lesion, root canal treatment, IA-guided endodontics, pulpectomy, in English; combined with each other using the Boolean operators “OR” and “AND”. The inclusion criteria are: articles published in the last five years (2019–2024), articles that address the different perspectives of the use of artificial intelligence in endodontics, articles that analyze the relationship between AI and endodontics, articles published after the start of the use of AI in endodontics. Articles that do not meet the defined inclusion criteria will be excluded. The exclusion criteria are: duplicate articles, articles requiring additional costs for access without full-text availability, and studies that are unclear and unexplained. This literature review aims to synthesize the findings related to artificial intelligence in endodontic practice over the past five years, thus creating a valuable source of information for healthcare professionals seeking applications and perspectives of artificial intelligence in endodontics.
Artificial intelligence has experienced exponential growth in recent years and is beginning to be used in various areas of dentistry, particularly in endodontics. In the field of endodontics, it has significantly evolved to assist in the diagnosis of pulp infections, based on radiographic analyses or dental scanners. Radiographic recognition AI programs use large datasets to establish their diagnoses and predict the progression of periapical diseases. The databases used by AI software are continuously improved and expanded as they are used, becoming increasingly precise. Another relevant aspect is the possibility, through artificial intelligence, of optimizing treatments thanks to systems capable of creating personalized treatment plans adapted to each patient based on their own anatomical characteristics. Artificial intelligence is still in its early stages but already enables the detection of periapical lesions, intracanal file fragments, bone loss, dental malformations, previously performed treatments, carious lesions, anatomical structures, etc. Some software can already predict the progression of dental diseases and the potential need for treatments. The objective of this narrative review is to evaluate the relevance and impact of the potential use of artificial intelligence in endodontics by identifying the objectives, various perspectives, and limits of its use, enabling accurate diagnosis and treatment planning and, thus, more effective endodontic practice. The bibliographic research will be conducted using search engines such as PubMed, Science Direct/Springer Link, B-on, and conference and congress abstract summaries. During the research, various MeSH (Medical Subject Headings) terms corresponding to the following keywords were used: Artificial intelligence/IA/AI, endodontics, machine learning, Convolutional Neural Network/CNN, apical lesion, root canal treatment, IA-guided endodontics, pulpectomy, in English; combined with each other using the Boolean operators “OR” and “AND”. The inclusion criteria are: articles published in the last five years (2019–2024), articles that address the different perspectives of the use of artificial intelligence in endodontics, articles that analyze the relationship between AI and endodontics, articles published after the start of the use of AI in endodontics. Articles that do not meet the defined inclusion criteria will be excluded. The exclusion criteria are: duplicate articles, articles requiring additional costs for access without full-text availability, and studies that are unclear and unexplained. This literature review aims to synthesize the findings related to artificial intelligence in endodontic practice over the past five years, thus creating a valuable source of information for healthcare professionals seeking applications and perspectives of artificial intelligence in endodontics.
Description
Keywords
Inteligência artificial Endodontia CNN ML Tratamento endodôntico Artificial intelligence Endodontics Endodontic treatment