| Grant number: | 22/03123-5 |
| Support Opportunities: | Scholarships in Brazil - Doctorate |
| Start date: | December 01, 2022 |
| Status: | Discontinued |
| Field of knowledge: | Health Sciences - Dentistry |
| Principal Investigator: | Pablo Agustin Vargas |
| Grantee: | Lucas Lacerda de Souza |
| Host Institution: | Faculdade de Odontologia de Piracicaba (FOP). Universidade Estadual de Campinas (UNICAMP). Piracicaba , SP, Brazil |
| Associated scholarship(s): | 23/11058-1 - Artificial intelligence in the diagnosis of low-grade and high-grade head and neck lymphomas., BE.EP.DR |
Abstract Small, round and blue cell neoplasms are tumors that represent a challenge to pathologists, as they mainly share similar morphological, immunophenotypic and molecular characteristics. With the advent of more advanced computational technologies, the use of artificial intelligence techniques through digital systems and the development of algorithms for image analysis have been studied to improve agility in diagnosis. This study aims to develop a system based on Artificial Intelligence to differentiate small, round and blue cell neoplasms through a Deep Learning approach, in order to diagnose lesions with superimposed microscopic characteristics that represent challenges in the differential diagnosis of the pathology. For this, samples that present the diagnosis of the group of lesions in question will be retrospectively retrieved. The slides will be scanned, and manual annotations will be made for the validation of the images, seeking delimitation of lesional tissue and normal tissue. The images will be divided and will guide the "labels" of the patches generated after their fragmentation, being resized to a fixed size. A random division of the images will be performed using a code network according to the test that is sought to be carried out in the neural network. After the validation of neural network learning, the deep learning methodology will be used, generating a classification of the characteristics that differentiate the analyzed groups. At the end of this experiment, it is expected to obtain an algorithm that supports an assertive histopathological diagnosis for a better management of the patient. | |
| News published in Agência FAPESP Newsletter about the scholarship: | |
| More itemsLess items | |
| TITULO | |
| Articles published in other media outlets ( ): | |
| More itemsLess items | |
| VEICULO: TITULO (DATA) | |
| VEICULO: TITULO (DATA) | |