Scope of Consultancy Services:

• Enhance the team’s technology ability on machine learning (ML), deep learning (DL), computer vision (CV) and medical image processing.
• Provide project service, help to solve the technique issues encountered during the project and instruct on algorithm model improvement.
• Consultancy on the medical image computer aided diagnosis (CAD) related data augmentation, generalized deep learning model design, medical image analysis etc. industry direction and academic research scope.
• Bi-weekly meeting with the team.

Deliverables:

• 1st Project Milestone
• Provide at least 2 technical lectures related to intelligent medical image analysis:

1. Technology trend on machine learning/deep learning for medical image analysis
2. A systematic review on data argumentation techniques

• Offer guidance and advice to the DL/ML algorithm design, which includes:

1. Generative Adversarial Network (GAN) model design to address data imbalance and sparsity issue for endoscopic image analysis.
2. Self-Supervised Learning (SSL) methods to solve the problem of low quality and low quantity of data labelling.
3. Generalized DL model design for endoscopic image analysis.

• 2^nd Project Milestone
• Provide at least 1 technical lecture related to intelligent medical image analysis:

1. A systematic review on the self- and semi-supervised learning

• Review the algorithm models and results, instruct on the performance improvement, which includes:

1. Lesion detection and early gastric cancer identification in endoscopic image
2. Endoscopic image anatomical site classification

• Contribute one innovative idea in the field of AI medical image analysis.

Requirements

• Ph.D. Degree in Computer Science or related disciplines
• 5+ years of experience in ML/DL algorithm research or computer vision related research in the field of medical image analysis.
• Both industry and academic research experience are highly preferred; can understand the requirement, formulate the problem, and propose innovative solutions addressing the problem effectively
• Good understanding and appreciation of algorithmic level fine tuning versus system-level performance optimization – various trade-off, parameters, adjustment, and modelling effects etc.
• Broad knowledge and connection in the community – keep abreast of the state of art of the latest progress, other industry players approach and academic R&D direction.