The role of Magnetic Resonance Imaging (MRI) in medicine and science has been well established, especially in neuroimaging. Access to MRI ranges from being prohibitive to scarcely available in developing countries. This is a critical gap because approximately eighty percent of the world’s population and ninety-seven percent of the population growth will be in developing countries. For example, Sub-Saharan Africa suffers a deficit of neurosurgeons (~1 per 7 million) and MRI scanners (0.04 per million). The lack of human expertise impacts neuroimaging of diseases like brain tumors, stroke, and Dementia. An important component of the challenge is the absence of skilled manpower required for MRI operation. The lack of educational facilities and the high costs in imparting technical training result in a lack of skilled manpower required to operate and utilize MRI systems. In developed countries, Inefficient workflows result in challenges related to financial and temporal access. Therefore, there is a need to augment human expertise to improve access, tackle “protocol creep” and standardize the practice.
In this Research Topic, we aim to explore and investigate the ability of machine learning (ML) methods to augment human MR expertise required for neuroimaging investigations. This research overview will be accomplished by collating research reports, mini-reviews, and data reports related to the use of ML methods to automate operation and enable interpretation of neuropathologies such as Hydrocephalus, stroke, brain tumors, and Dementia. In particular, the Research Topic will engage in defining levels of autonomous MRI similar to the six levels of autonomy for automobiles. The Research Topic will focus on the latest developments in automated neuro-image acquisition, analysis, quality assurance, patient monitoring, subject setup, and deployment. These developments will be discussed in the clinical and scientific contexts by end-users of such automated methods to enhance neuroimaging practice. In summary, the Research Topic will embark on the first steps to disseminate the opportunities and challenges of autonomous MRI for brain imaging.
The subtopics will be categorized based on the preliminary work reported in Ravi and Geethanath, MRI 2020 detailing the autonomous MRI for the first time. The subtopics will be classified into the six levels of autonomy similar to the classification by the society of Automobile Engineers (SAE) for autonomous vehicles. The list of potential subtopics include:
1. Autonomous MRI of the brain: Why, what, and how?
2. Autonomous MRI: Opportunities and challenges
3. Automating image acquisition
4. Automating image quality assessment
5. Automating patient monitoring and engagement
6. Virtual scanner and the MRI digital twin
7. Automating Hydrocephalus detection
8. Automating Dementia screening
9. Automating Stroke detection
10. Automating Brain tumors detection and follow up
Dr. Geethanath is the co-founder of MR Access, the inventor of multiple US patent applications and he received funding from GE Healthcare, USA, and GE Healthcare, India. The other Topic Editors declare no competing interests with the topic.
The role of Magnetic Resonance Imaging (MRI) in medicine and science has been well established, especially in neuroimaging. Access to MRI ranges from being prohibitive to scarcely available in developing countries. This is a critical gap because approximately eighty percent of the world’s population and ninety-seven percent of the population growth will be in developing countries. For example, Sub-Saharan Africa suffers a deficit of neurosurgeons (~1 per 7 million) and MRI scanners (0.04 per million). The lack of human expertise impacts neuroimaging of diseases like brain tumors, stroke, and Dementia. An important component of the challenge is the absence of skilled manpower required for MRI operation. The lack of educational facilities and the high costs in imparting technical training result in a lack of skilled manpower required to operate and utilize MRI systems. In developed countries, Inefficient workflows result in challenges related to financial and temporal access. Therefore, there is a need to augment human expertise to improve access, tackle “protocol creep” and standardize the practice.
In this Research Topic, we aim to explore and investigate the ability of machine learning (ML) methods to augment human MR expertise required for neuroimaging investigations. This research overview will be accomplished by collating research reports, mini-reviews, and data reports related to the use of ML methods to automate operation and enable interpretation of neuropathologies such as Hydrocephalus, stroke, brain tumors, and Dementia. In particular, the Research Topic will engage in defining levels of autonomous MRI similar to the six levels of autonomy for automobiles. The Research Topic will focus on the latest developments in automated neuro-image acquisition, analysis, quality assurance, patient monitoring, subject setup, and deployment. These developments will be discussed in the clinical and scientific contexts by end-users of such automated methods to enhance neuroimaging practice. In summary, the Research Topic will embark on the first steps to disseminate the opportunities and challenges of autonomous MRI for brain imaging.
The subtopics will be categorized based on the preliminary work reported in Ravi and Geethanath, MRI 2020 detailing the autonomous MRI for the first time. The subtopics will be classified into the six levels of autonomy similar to the classification by the society of Automobile Engineers (SAE) for autonomous vehicles. The list of potential subtopics include:
1. Autonomous MRI of the brain: Why, what, and how?
2. Autonomous MRI: Opportunities and challenges
3. Automating image acquisition
4. Automating image quality assessment
5. Automating patient monitoring and engagement
6. Virtual scanner and the MRI digital twin
7. Automating Hydrocephalus detection
8. Automating Dementia screening
9. Automating Stroke detection
10. Automating Brain tumors detection and follow up
Dr. Geethanath is the co-founder of MR Access, the inventor of multiple US patent applications and he received funding from GE Healthcare, USA, and GE Healthcare, India. The other Topic Editors declare no competing interests with the topic.