Adding better graphical results for model training and deployments

[mxochicale]

Unfortunately, Figure 2 only shows training for 5 subjects and results from 31 subjects need to be added. Also, results can be improved showing perhaps some curves per model:

[mxochicale]

L. Pisani To Everyone 10:26:31 AM LP A FAIRsharing compliant CSV file containing the census data used in this evaluation is available at the following address (https://doi.org/10.6084/m9.figshare.15167406.v1)..) For further information regarding the data and the CCA, please contact the CCA data access committee (DAC@nicslk.com) and quote the manuscript, your institution and provide return correspondence information. This is an example of what we are using in manuscripts, using figshare

[mxochicale]

• [x] Missing files in /home/mx19/repositories/vital-ultrasound/ML4H2022/data/results/aug-09-2022_SqueezeNet_hyperparameters/frames_per_clip/05-subjects:

  # FRAMES_PER_SEGMENT_IN_A_CLIP = 'NUMBER_OF_FRAMES_PER_SEGMENT_IN_A_CLIP_30' REPRODUCED
  # FRAMES_PER_SEGMENT_IN_A_CLIP = 'NUMBER_OF_FRAMES_PER_SEGMENT_IN_A_CLIP_60' LEAVE FOR OTHER TO REPRODUCE

[mxochicale]

Other relevant references

Shah, Pratik, Francis Kendall, Sean Khozin, Ryan Goosen, Jianying Hu, Jason Laramie, Michael Ringel, and Nicholas Schork. "Artificial intelligence and machine learning in clinical development: a translational perspective." NPJ digital medicine 2, no. 1 (2019): 1-5. https://scholar.google.com/scholar?cites=11957967473536523492&as_sdt=2005&sciodt=0,5&hl=en

[mxochicale]

Data Access Process

Robin Carpenter, Ethics & Governance Lead AI Centre, King's College London, presenting Data Access Process at What does robust governance of a learning healthcare system look like? YouTube Video

[mxochicale]

Close it due to limited bandwidth but feel free to open it for future work and discussions.