• Tomorrow's Dose
  • Posts
  • Edition 1 - AI for personalised cancer treatment, deep-learning tumour probability maps and more

Edition 1 - AI for personalised cancer treatment, deep-learning tumour probability maps and more

Lots of new developments, including, AI certificate course and Whole-body, ultralow-field 0.05 T MR scanner

July 24, 2024

  1. AI utilised for personalised cancer treatment strategies.

  2. Clinicians preferred deep-learning tumour probability maps over a single-contour output.

  3. An AI certificate course can improve knowledge of the technology for radiology residents.

  4. Whole-body, ultralow-field 0.05 T MR scanner that operates from a standard wall power outlet.

Specialty: Oncology // Sub-Specialty: AI // Body Site: Prostate


1. AI utilised for personalised cancer treatment strategies:


Applied deep reinforcement learning has been utilised to create adaptive therapy schedules for prostate cancer. The personalised approach significantly outperformed standard-of-care treatments and existing adaptive therapy approaches.
Read Full Article

Paul’s Thoughts:

The holy grail of oncology treatment is to deliver personalised medicine, and it seems that with AI we may have a powerful tool that will allow this to happen.

Timescale: Mid | 5 Years 

Specialty: Radiotherapy // Sub-Specialty: Segmentation // Body Site: All


2. Clinicians preferred deep-learning tumour probability maps over a single-contour output:


A graphical user interface (GUI) was designed to allow the user to interact with tumour probability maps. A study was conducted where nine experts in tumour delineation interacted with the interface prototype and its functionality. The participants’ experience was assessed qualitatively and quantitatively. Interviews with radiation oncologists revealed their preference for using a rainbow colour map to visualise tumour probability maps during contouring, which they found intuitive.
Read Full Article

Paul’s Thoughts:

Deep learning segmentation is already commonplace in radiotherapy clinics worldwide, the uncertainty in the contours is currently missing in commercial products. This study shows that clinicians prefer to see a tumour probability map as this will guide them as to which parts of the contour require the most attention.

Timescale: Early | 2 Years 

Specialty: Radiotherapy // Sub-Specialty: Education // Body Site: All


3. An AI certificate course can improve knowledge of the technology for radiology residents:


A pilot study including 42 radiology residents at two separate residency programs participated in the RSNA Imaging AI Foundational Certificate course over a four-month period. The team reported a precourse assessment score of 37%, which increased significantly to 73% after the modules were completed (p < 0.001). Also, 31 of the participating residents (74%) agreed that the coursework improved their familiarity with AI in radiology.
Read Full Article

Paul’s Thoughts:

This study is vital as it bridges the knowledge gap that currently exists regarding the introduction and integration of AI education in radiology residency programs.

Timescale: Early | 3 Years 

Specialty: Radiology // Sub-Specialty: Equipment // Body Site: All


4. Whole-body, ultralow-field 0.05 T MR scanner that operates from a standard wall power outlet:


Along with deep-learning reconstruction methods, the scanner was able to obtain anatomical images of the brain, spine, abdomen and knee with comparable image quality to that of a 3T system.
Read Full Article

Paul’s Thoughts:

This is a fascinating development, with the Organisation for Economic Co-operation and Development (OCED) reporting that the access to MRI scanners is extremely heterogeneous: the number of MRI scanners (in 2021) ranged from just 0.24 per million people in Columbia to 55 per million in Japan.

Timescale:  Mid | 5 Years 

A round-up of some of the best posts we found online this week.

REGULATION
Clinicians don't really care about healthtech regulations

Dr Zhong Wei Khor | NHS Doctor | The Beatson Oncology Center

EMPLOYMENT
Is it Silicon Valley's jobs to make guaranteed income a reality?

Rudolf Wagner | AI Compliance Officer | Freelance

EMPLOYMENT
Will AI lead to new jobs in healthcare?

Jan Beger | Head of AI Advocacy | GE Healthcare

POLICY
NHS Cancer care is at a tipping point

Ajay Aggarwal | Professor and Consultant Clinical Oncologist | London School of Hygiene and Tropical Medicine

Was this email forwarded to you?
Our weekly email brings you the latest health trends and insights, combining top news and opinions into a straightforward, digestible format.

Meet The Editor

Want an article featured?

Have an insightful link or story about the future of medical health? Reach out below, and we may include it in a future release.

Reply

or to participate.