• Corpus ID: 237428138

Utilizing a digital swarm intelligence platform to improve consensus among radiologists and exploring its applications

@inproceedings{Shah2021UtilizingAD,
  title={Utilizing a digital swarm intelligence platform to improve consensus among radiologists and exploring its applications},
  author={Rutwik Shah and Bruno Astuto and Tyler J. Gleason and William Fletcher and J Banaga and Kevin Sweetwood and Allen Ye and Rina P. Patel and Kevin C McGill and Thomas M. Link and Jason C. Crane and Valentina Pedoia and Sharmila Majumdar},
  year={2021}
}
Introduction: Radiologists today play a central role in making diagnostic decisions and labeling images for training and benchmarking Artificial Intelligence (A.I.) algorithms. A key concern is low inter-reader reliability (IRR) seen between experts when interpreting challenging cases. While teams-based decisions are known to outperform individual decisions, inter-personal biases often creep up in group interactions which limit non-dominant participants from expressing true opinions. To… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 49 REFERENCES
Human–machine partnership with artificial intelligence for chest radiograph diagnosis
TLDR
The superior diagnostic accuracy of the combined HITL AI solution compared to radiologists and AI alone has broad implications for the surging clinical AI deployment and implementation strategies in future practice.
Boosting medical diagnostics by pooling independent judgments
TLDR
It is found that similarity in diagnostic accuracy is a key condition for collective intelligence: Aggregating the independent judgments of doctors outperforms the best doctor in a group whenever the diagnostic accuracy of doctors is relatively similar, but not when doctors’ diagnostic accuracy differs too much.
Automatic Deep Learning-assisted Detection and Grading of Abnormalities in Knee MRI Studies.
TLDR
Three-dimensional convolutional neural network had high sensitivity, specificity, and accuracy for knee-lesion-severity scoring and also increased intergrader agreement when used on an external dataset.
A User Interface for Optimizing Radiologist Engagement in Image Data Curation for Artificial Intelligence.
TLDR
GUI-component compatibility with common image analysis tools facilitates radiologist engagement in image data curation, including image annotation, supporting AI application development and evolution for medical imaging.
Radiologic criteria for the diagnosis of spinal stenosis: results of a Delphi survey.
TLDR
There are no broadly accepted quantitative criteria and only partially accepted qualitative criteria for the diagnosis of lumbar spinal stenosis, which include disk protrusion, lack of perineural intraforaminal fat, hypertrophic facet joint degeneration, absent fluid around the cauda equine, and hypertrophy of the ligamentum flavum.
Mammographic Breast Density Assessment Using Deep Learning: Clinical Implementation.
TLDR
This DL model can be used to assess mammographic breast density at the level of an experienced mammographer and was implemented in routine clinical practice, where eight radiologists reviewed 10 763 consecutive mammograms assessed with the model.
AggNet: Deep Learning From Crowds for Mitosis Detection in Breast Cancer Histology Images
TLDR
An experimental study on learning from crowds that handles data aggregation directly as part of the learning process of the convolutional neural network (CNN) via additional crowdsourcing layer (AggNet), which gives valuable insights into the functionality of deep CNN learning from crowd annotations and proves the necessity of data aggregation integration.
The Wisdom of Crowds of Doctors
TLDR
This study suggests that the mean of the clinicians’ predictions can quickly approach the accuracy of the best clinician using as few as 5 clinicians, and averaging clinician predictions could prove helpful in situations where statistical models do not yet exist or where existing models are inadequate.
Inter- and intrareader agreement for categorization of background parenchymal enhancement at baseline and after training.
TLDR
Initial interreader agreement for BPE was fair among breast radiologists but achieved sustained improvement with training, highlighting the importance of education and inclusion of standardized BPE categories in a reference atlas.
...
...