Human Artificial Intelligence Loop (HAIL) » Computational Microscopy Imaging Laboratory » College of Medicine

In this work, we have developed a tool that uses AI to annotate large images of histologically stained tissues and display AI predictions in a commonly used digital pathology whole slide viewer. Users can employ the native whole slide annotation tool to label structures in whole slide images (WSIs) both to train AI models and also to correct the predictions made on new samples as feedback for the network. This correction process is repeated until the desired performance is achieved.

**HAIL for histological segmentation of ROIs.** (A) HAIL iteratively reduces the annotation time required to generate accurate segmentations. (B) Multi-resolution application of HAIL improves performance over single resolution. (C) Segmentation of glomerular boundaries from a WSI, using HAIL.

Select References

B. Lutnick, B. Ginley, D. Govind, S. D. McGarry, P. S. LaViolette, R. Yacoub, S. Jain, J. E. Tomaszewski, K. Y. Jen, and P. Sarder, “An integrated iterative annotation technique for easing neural network training in medical image analysis,” Nature Machine Intelligence, vol. 1, no. 2, pp. 112–19, Feb. 2019.

Underline indicates corresponding author.
^* indicates equal contribution.
^Ω indicates Dr. Sarder’s faculty trainees.
^§ indicates post doctoral associates.
^† indicates Dr. Sarder’s graduate students.
^‡ indicates Dr. Sarder’s undergraduate students.