Titulo Estágio
Landmark detection in orthopedic implants
Áreas de especialidade
Sistemas Inteligentes
Sistemas Inteligentes
Local do Estágio
Full remote, with optional office space in Braga.
Enquadramento
In the preoperative planning process, it is common practice to place anatomical landmarks in the medical images. This process can be done either manually or automatically, using Machine or Deep learning techniques.When the landmarks in the medical images are present, it is possible to automate a series of planning steps, such as the determination of procedure-specific measurements or proper implant placement. A less touched upon but necessary step in the process of automating the implant placement is the landmark and axis definition in the implant itself. For the coupling between bone and implant to occur, the 3D representations of the implants must also have landmark information. This process can be done manually but it is a tedious, labor-intensive task, which can be replaced with automatic processes.For that purpose, a Deep Learning model can be trained to detect these landmarks on the 3D models of a given implant type, such as femoral knee replacement components.
Objetivo
Dataset:
Organize and prepare data for project
Exploratory data analysis
Literature review:
Compile a comprehensive literature review on domain specific themes (surgical planning, types of knee implants…) as well as technical relevant approaches
Modeling:
Implement and train a Deep Learning Model for the automatic detection of anatomical landmarks in femoral knee replacement implants
Create and iterate over the data processing-model training-model evaluation pipeline
Evaluation:
Implement relevant metrics for the problem assessment
Design and conduct qualitative tests with domain experts
Application (optional):
Create an interactive tool for application of the developed model
Plano de Trabalhos - Semestre 1
Organize and prepare data: 9/15/2022 to 10/15/2022
Exploratory data analysis: 10/15/2022 to 10/30/2022
Background Research: 11/1/2022 to 11/16/2022
Literature Review: 11/1/2022 to 12/1/2022
Write Intermediate Thesis Report: 12/1/2022 to 12/31/2022
Prepare Intermediate Thesis Presentation: 1/1/2023 to 1/16/2023
Plano de Trabalhos - Semestre 2
Implement and train a DL Model: 2/15/2023 to 3/17/2023
Implement numerical result evaluation: 3/1/2023 to 3/16/2023
Implement qualitative result evaluation with domain experts: 4/15/2023 to 5/15/2023
Iterate over the data-model-evaluation process: 3/15/2023 to 4/14/2023
Create web service for model application (optional): 5/15/2023 to 6/14/2023
Write Thesis: 5/15/2023 to 7/14/2023
Prepare Thesis Presentation: 7/1/2023 to 7/16/2023
Condições
---
Observações
For their work, the student will use internal company datasets.
Orientador
Ana Costa
📩