Six must-have features for advanced analytics and visualization platform for validating new-age AI algorithms

Congress:

ECR 2020

Poster Number:

C-14851

Type:

Educational Exhibit

Keywords:

Artificial Intelligence and Machine Learning, Artificial Intelligence, CT, Digital radiography, MR, Comparative studies, Computer Applications-Detection, diagnosis, Quality assurance, Not applicable

Authors:

R. Takhar, S. Gupta, S. Gaur, C. Kapoor, V. Mahajan, V. Venugopal, H. Mahajan ; New Delhi/IN

DOI:

10.26044/ecr2020/C-14851

DOI-Link:

https://dx.doi.org/10.26044/ecr2020/C-14851

Fig. 1: Scatter-plot on CARPL depicting the distribution of cases with matched and...

Fig. 2: Dot Plot displaying a comparison of results between radiologists and an AI...

Fig. 3: Visualisation for an AI algorithm that quantifies spinal canal dimensions on...

Fig. 4: Arbitration window on 'CARPL' platform where the results of AI and radiologist...

Learning objectives

In this exhibit, we discuss the six key must-have features on any analytical platform that is intended for the validation of AI algorithms. With the recent developments in machine learning and especially deep learning, a lot of companies are trying to develop solutions for assisting radiologists in medical imaging. We have developed a system that combines statistics with medical inputs to provide insights and validate deep learning algorithms at scale. One of the key challenges is the variety of output from these algorithms. The output...

Background

We believe that the usage of such tools will decrease the time required for validating the deep learning algorithms in healthcare setup and at the same time, will provide useful insights to the companies which will help them in improving the algorithm further.

Findings and procedure details

Ability to fetch data from PACS: To conduct a study, the hospital/clinic should be able to easily search and extract cases. The tool should have features to filter cases on the basis of modalities, diseases, and other related fields. In addition to these, advanced features like semantic search can be really useful to capture the diversity of diseases and modalities. Our system provides features to include/exclude modalities and diseases. Client-side anonymization: Data privacy and security is an important aspect of any validation study. The system...

Conclusion

Once the arbitration process is done, the system should generate a summary of the algorithm’s performance on a set of parameters. These parameters can vary depending on the type of modality and the study.

Personal information and conflict of interest

V. K. Venugopal; New Delhi/IN - nothing to disclose S. Gupta; New Delhi/IN - nothing to disclose V. Mahajan; New Delhi/IN - nothing to disclose S. Gaur; New Delhi/IN - nothing to disclose C. Kapoor; New Delhi/IN - nothing to disclose H. Mahajan; New Delhi/IN - nothing to disclose R. Takhar; New Delhi/IN - nothing to disclose

References

1.Yepes-Calderon, F., Nelson, M. D. & McComb, J. G. Automatically measuring brain ventricular volume within PACS using artificial intelligence. PLoS ONE 13, e0193152 (2018). 2.Tang, A. et al. Canadian Association of Radiologists White Paper on Artificial Intelligence in Radiology. Can Assoc Radiol J 69, 120–135 (2018). 3.Savadjiev, P. et al. Demystification of AI-driven medical image interpretation: past, present and future. Eur Radiol 29, 1616–1624 (2019). 4.Thrall, J. H. et al. Artificial Intelligence and Machine Learning in Radiology: Opportunities, Challenges, Pitfalls, and Criteria for Success. J...