HistoLens » Computational Microscopy Imaging Laboratory » College of Medicine

HistoLens is a standalone desktop software for quantitative evaluation of annotated structures in histological whole slide images (WSIs). Users are able to interact with their data through dynamic labeling, interactive figure generation, visualization of hand-engineered features, training a variety of classification algorithms, and generating annotations of histological lesions for downstream analysis.

Screenshot of the Histolens2 main window. The window is divided into sections: Input Property Name with a field showing ‘Structure’ and a value of ‘Ts2P87’, Input Files with a scrollable list of files, Feature Rankings with a Similarity Threshold slider, and a large graph labeled ‘PCA 1’ and ‘PCA 2’ displaying points. Below the graph are labeling components options. The bottom of the window contains tabs labeled ‘Input files’, ‘Image viewing’, ‘Select features’ and ‘Analysis toolbox’. Below the tabs is a description reading: Input path goes to input extraction and result experiment file to reload previous analysis.

Resources

Select References

S. P. Border^†, B. Ginley^†, J. E. Tomaszewski, and P. Sarder, “HistoLens: a generalizable tool for increasing accessibility and interpretability of quantitative analyses in digital pathology,” Proc. of SPIE–Medical Imaging 2022: Digital & Computational Pathology, vol. 12039, pp. 1203929:1–8, San Diego, CA, USA, Feb. 2022.

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