Search

Objective: To investigate the performance of ChatGPT in the differential diagnosis of oral and maxillofacial diseases., Methods: Thirty-seven oral and maxillofacial lesions findings were presented to ChatGPT-3.5 and - 4, 18 dental surgeons trained in oral medicine/pathology (OMP), 23 general dental surgeons (DDS), and 16 dental students (DS) for differential diagnosis. Additionally, a group of 15 general dentists was asked to describe 11 cases to ChatGPT versions. The ChatGPT-3.5, -4, and human primary and alternative diagnoses were rated by 2 independent investigators with a 4 Likert-Scale. The consistency of ChatGPT-3.5 and - 4 was evaluated with regenerated inputs., Results: Moderate consistency of outputs was observed for ChatGPT-3.5 and - 4 to provide primary (κ = 0.532 and κ = 0.533 respectively) and alternative (κ = 0.337 and κ = 0.367 respectively) hypotheses. The mean of correct diagnoses was 64.86% for ChatGPT-3.5, 80.18% for ChatGPT-4, 86.64% for OMP, 24.32% for DDS, and 16.67% for DS. The mean correct primary hypothesis rates were 45.95% for ChatGPT-3.5, 61.80% for ChatGPT-4, 82.28% for OMP, 22.72% for DDS, and 15.77% for DS. The mean correct diagnosis rate for ChatGPT-3.5 with standard descriptions was 64.86%, compared to 45.95% with participants' descriptions. For ChatGPT-4, the mean was 80.18% with standard descriptions and 61.80% with participant descriptions., Conclusion: ChatGPT-4 demonstrates an accuracy comparable to specialists to provide differential diagnosis for oral and maxillofacial diseases. Consistency of ChatGPT to provide diagnostic hypotheses for oral diseases cases is moderate, representing a weakness for clinical application. The quality of case documentation and descriptions impacts significantly on the performance of ChatGPT., Clinical Relevance: General dentists, dental students and specialists in oral medicine and pathology may benefit from ChatGPT-4 as an auxiliary method to define differential diagnosis for oral and maxillofacial lesions, but its accuracy is dependent on precise case descriptions., (© 2024. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.)

The Zika virus (ZIKV) has shown a promising oncolytic effect against embryonal CNS tumors. However, studies on the effect of different administration routes and the ideal viral load in preclinical models are highly relevant aiming for treatment safety and efficiency. Here, we investigated the effect and effectiveness of different routes of administration, and the number of ZIKV BR injections on tumor tropism, destruction, and side effects. Furthermore, we designed an early-stage human brain organoid co-cultured with embryonal CNS tumors to analyze the ZIKV BR oncolytic effect. We showed that in the mice bearing subcutaneous tumors, the ZIKV BR systemically presented a tropism to the brain. When the tumor was located in the mice's brain, serial systemic injections presented efficient tumor destruction, with no neurological or other organ injury and increased mice survival. In the human cerebral organoid model co-cultured with embryonal CNS tumor cells, ZIKV BR impaired tumor progression. The gene expression of cytokines and chemokines in both models suggested an enhancement of immune cells recruitment and tumor inflammation after the treatment. These results open new perspectives for virotherapy using the ZIKV BR systemic administration route and multiple doses of low virus load for safe and effective treatment of embryonal CNS tumors, an orphan disease that urges new effective therapies.