A smartphone application to improve the precision of biopsy site identification: A proof-of-concept study.

Dermatologists rely on skin biopsies to diagnose cutaneous tumors and rashes. Skin biopsy sites should be accurately identified with conventional anatomical site descriptors in the pathology request form. Reliance upon free-text entries to describe these biopsy sites is prone to user error and can cause medical misadventures such as wrong-site follow-up surgery. We sought to determine whether a smartphone application (RightSite) could improve the precision of biopsy site labeling. We conducted a prospective proof-of-concept study of 100 smartphone-assisted skin biopsy site identifiers with matched comparison to 100 historical controls. Student's t-test was used to identify significant differences in the precision of anatomic descriptors before and after adoption of the application. We found a 69% improvement in precision of anatomic site labeling with the RightSite smartphone application (P < 0.0001). These data show smartphone-assisted biopsy site labeling improves the precision of anatomic site descriptors. Integrating graphical user interfaces into the electronic health records system could improve health care by standardizing anatomic site nomenclature and site-specific descriptors.

Automatic segmentation of intravital fluorescence microscopy images by K-means clustering of FLIM phasors.

Fluorescence lifetime imaging microscopy (FLIM) provides additional contrast for fluorophores with overlapping emission spectra. The phasor approach to FLIM greatly reduces the complexity of FLIM analysis and enables a useful image segmentation technique by selecting adjacent phasor points and labeling their corresponding pixels with different colors. This phasor labeling process, however, is empirical and could lead to biased results. In this Letter, we present a novel and unbiased approach to automate the phasor labeling process using an unsupervised machine learning technique, i.e., K-means clustering. In addition, we provide an open-source, user-friendly program that enables users to easily employ the proposed approach. We demonstrate successful image segmentation on 2D and 3D FLIM images of fixed cells and living animals acquired with two different FLIM systems. Finally, we evaluate how different parameters affect the segmentation result and provide a guideline for users to achieve optimal performance.