Harvard Catalyst Profiles

Biopsy and histopathological imaging of biopsied tissues stained with Hematoxylin and Eosin (H&E) remain the standard clinical method for diagnosis. Although this approach has been the gold standard for over a century, it can be considerably inaccurate. For instance, the U.S. national benchmark for the positive predictive value of malignancy in breast cancer biopsy assessments is only about 30%, meaning that 7 out of 10 biopsies return negative results for cancer. This high rate of false negatives can be attributed to two key factors: 1) the target tissue is often highly heterogeneous, leading to inaccurate biopsy sampling, and 2) histopathological imaging is typically performed off-site, which is time-consuming and does not provide real-time feedback for on-site decision-making. To address this issue, we propose the development of an "in-situ histopathological lab" capable of capturing in-vivo images of target tissues without the need for a biopsy. Toward this goal, we have developed the first-of-its-kind bendable graded index (bGRIN) lens for distal-scan-free 3D microendoscopy, marking a significant step toward realizing a practical "in-situ histopathological lab". This system currently integrates three imaging modalities, i.e., two-photon imaging, second harmonic imaging, as well as photoacoustic imaging, to acquire images with more comprehensive information than H&E images. The students will have the opportunity to get involved in the further development of this promising cutting-edge medical imaging technology. Specifically, training on various optical imaging modalities, hardware and system development, image processing and analysis, as well as summarizing findings in biology and medicine is expected. After initial training, the students may play a significant and independent role in the project through freely generating and implementing ideas, with strong support and guidance from the mentor, colleagues, and collaborators.