APEHR: Automated Prognosis in Electronic Health Records using multi-head self-attention

Autor(es):Mostrar menos -

Florez, Alexander Y. C. ; Scabora, Lucas ; Eler, Danilo M. ; Rodrigues-Jr, Jose F. ; Almeida, JR ; Gonzalez, AR ; Shen, L ; Kane, B ; Traina, A ; Soda, P ; Oliveira, JL

• Tipo de documento: Artigo Científico
• Fonte: 2021 IEEE 34TH INTERNATIONAL SYMPOSIUM ON COMPUTER-BASED MEDICAL SYSTEMS (CBMS); v. N/A, p. 6-pg., 2021-01-01.

Resumo

Automated prognosis has been a topic of intense research. Many works have sought to learn from Electronic Health Records using Recurrent Neural Networks that, despite promising results, have been overcome by novel techniques. We introduce APEHR, a Transformer approach that leverages medical prognosis using the latest technology Neural Network Transformer, which has demonstrated superior results in problems whose data is organized in sequential fashion. We contribute with an innovative problem modeling along with a detailed discussion of how Transformers can be used in the medical domain. Our results demonstrate a prognostic performance that surpasses previous works by at least 6% for metric Recall@k in the public dataset MIMIC-III. (AU)

• Processo FAPESP: 16/17330-1 - Armazenamento e Operações de Navegação em Grafos em SGBDs Relacionais
• Beneficiário: Lucas de Carvalho Scabora
• Modalidade de apoio: Bolsas no Brasil - Doutorado
• Processo FAPESP: 18/17620-5 - Medicina preventiva por meio de técnicas de deep learning aplicadas ao prognóstico de saúde
• Beneficiário: José Fernando Rodrigues Júnior
• Modalidade de apoio: Bolsas no Exterior - Pesquisa
• Processo FAPESP: 16/17078-0 - Mineração, indexação e visualização de Big Data no contexto de sistemas de apoio à decisão clínica (MIVisBD)
• Beneficiário: Agma Juci Machado Traina
• Modalidade de apoio: Auxílio à Pesquisa - Temático