Active Vision in Ultrasound Medical Imaging
Lead Research Organisation:
Imperial College London
Department Name: Computing
Abstract
The proposed project focuses on portable ultrasound scanners which aim to
be marketed to expert and amateur operators alike and send the images acquired
to expert medical practitioners for diagnosis. Thus the need arises for guiding the
operator to the correct point of view that will be necessary for accurate automatic
3D anatomy reconstruction in addition to helping physicians in making a better
diagnosis.
Medical Imaging
be marketed to expert and amateur operators alike and send the images acquired
to expert medical practitioners for diagnosis. Thus the need arises for guiding the
operator to the correct point of view that will be necessary for accurate automatic
3D anatomy reconstruction in addition to helping physicians in making a better
diagnosis.
Medical Imaging
Organisations
People |
ORCID iD |
Bernhard Kainz (Primary Supervisor) | |
Athanasios Vlontzos (Student) |
Publications
Vlontzos Athanasios
(2020)
Causal Future Prediction in a Minkowski Space-Time
in arXiv e-prints
Vlontzos Athanasios
(2019)
Multiple Landmark Detection using Multi-Agent Reinforcement Learning
in arXiv e-prints
Vlontzos Athanasios
(2020)
3D Probabilistic Segmentation and Volumetry from 2D projection images
in arXiv e-prints
Liu Tianrui
(2020)
Ultrasound Video Summarization using Deep Reinforcement Learning
in arXiv e-prints
Hou Benjamin
(2019)
Flexible Conditional Image Generation of Missing Data with Learned Mental Maps
in arXiv e-prints
Alansary A
(2019)
Evaluating reinforcement learning agents for anatomical landmark detection.
in Medical image analysis
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
EP/R513052/1 | 01/10/2018 | 30/09/2023 | |||
2130794 | Studentship | EP/R513052/1 | 01/10/2018 | 30/09/2022 | Athanasios Vlontzos |
Description | The purpose of this project is to help computer systems be able to navigate through and reason about the human body and its pathologies using medical image volumes. Medical images could include MRI, CT-Scans, Ultrasounds. We have shown that the knowledge of anatomy and physics aid computer agents perform the above mentioned actions more accurately |
Exploitation Route | Future researchers could expand on this work by implementing the outcomes of this project to real life scenarios in medical practice. |
Sectors | Digital/Communication/Information Technologies (including Software),Healthcare |