1
|
Combalia M, Garcia S, Malvehy J, Puig S, Mülberger AG, Browning J, Garcet S, Krueger JG, Lish SR, Lax R, Ren J, Stevenson M, Doudican N, Carucci JA, Jain M, White K, Rakos J, Gareau DS. Deep learning automated pathology in ex vivo microscopy. Biomed Opt Express 2021; 12:3103-3116. [PMID: 34221648 PMCID: PMC8221965 DOI: 10.1364/boe.422168] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/09/2021] [Revised: 03/29/2021] [Accepted: 03/30/2021] [Indexed: 05/09/2023]
Abstract
Standard histopathology is currently the gold standard for assessment of margin status in Mohs surgical removal of skin cancer. Ex vivo confocal microscopy (XVM) is potentially faster, less costly and inherently 3D/digital compared to standard histopathology. Despite these advantages, XVM use is not widespread due, in part, to the need for pathologists to retrain to interpret XVM images. We developed artificial intelligence (AI)-driven XVM pathology by implementing algorithms that render intuitive XVM pathology images identical to standard histopathology and produce automated tumor positivity maps. XVM images have fluorescence labeling of cellular and nuclear biology on the background of endogenous (unstained) reflectance contrast as a grounding counter-contrast. XVM images of 26 surgical excision specimens discarded after Mohs micrographic surgery were used to develop an XVM data pipeline with 4 stages: flattening, colorizing, enhancement and automated diagnosis. The first two stages were novel, deterministic image processing algorithms, and the second two were AI algorithms. Diagnostic sensitivity and specificity were calculated for basal cell carcinoma detection as proof of principal for the XVM image processing pipeline. The resulting diagnostic readouts mimicked the appearance of histopathology and found tumor positivity that required first collapsing the confocal stack to a 2D image optimized for cellular fluorescence contrast, then a dark field-to-bright field colorizing transformation, then either an AI image transformation for visual inspection or an AI diagnostic binary image segmentation of tumor obtaining a diagnostic sensitivity and specificity of 88% and 91% respectively. These results show that video-assisted micrographic XVM pathology could feasibly aid margin status determination in micrographic surgery of skin cancer.
Collapse
Affiliation(s)
- Marc Combalia
- Department of Dermatology, Hospital Clinic de Barcelona, Universitat de Barcelona, Barcelona, Spain
| | - Sergio Garcia
- Department of Dermatology, Hospital Clinic de Barcelona, Universitat de Barcelona, Barcelona, Spain
| | - Josep Malvehy
- Department of Dermatology, Hospital Clinic de Barcelona, Universitat de Barcelona, Barcelona, Spain
| | - Susana Puig
- Department of Dermatology, Hospital Clinic de Barcelona, Universitat de Barcelona, Barcelona, Spain
| | | | - James Browning
- The Rockefeller University, 1230 York Avenue, New York, NY 10065, USA
| | - Sandra Garcet
- The Rockefeller University, 1230 York Avenue, New York, NY 10065, USA
| | - James G. Krueger
- The Rockefeller University, 1230 York Avenue, New York, NY 10065, USA
| | - Samantha R. Lish
- The Rockefeller University, 1230 York Avenue, New York, NY 10065, USA
| | - Rivka Lax
- The Rockefeller University, 1230 York Avenue, New York, NY 10065, USA
| | - Jeannie Ren
- The Rockefeller University, 1230 York Avenue, New York, NY 10065, USA
| | - Mary Stevenson
- Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY 10065, USA
| | - Nicole Doudican
- Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY 10065, USA
| | - John A. Carucci
- Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY 10065, USA
| | - Manu Jain
- Ronald O. Pearlman Department of Dermatology, New York University, 550 First Avenue, New York, NY 10016, USA
| | - Kevin White
- Department of Dermatology, Oregon Health & Science University, 3303 South Bond Avenue, Portland, OR 97239, USA
| | - Jaroslav Rakos
- SurgiVance Inc., 310 East 67th Street, New York, NY 10065, USA
| | - Daniel S. Gareau
- The Rockefeller University, 1230 York Avenue, New York, NY 10065, USA
- SurgiVance Inc., 310 East 67th Street, New York, NY 10065, USA
| |
Collapse
|
2
|
Abstract
BACKGROUND Mohs micrographic surgery (MMS) is a frequently used technique that provides total margin visualization for treatment of skin neoplasms. OBJECTIVE To provide a comprehensive review of MMS literature, focusing on its origins, evidence behind present-day uses of MMS, and future directions. METHODS A literature search was conducted using PubMed to identify articles pertaining to MMS. RESULTS The fresh frozen technique led to widespread use of MMS in the 1970s. One randomized controlled trial and several large prospective studies have demonstrated low recurrence rates for treatment of nonmelanoma skin cancer (NMSC). MMS, when compared with surgical excision, also achieved a statistically significant higher cure rate for treatment of recurrent NMSC. Studies have demonstrated low recurrence for the treatment of melanoma and melanoma in situ with MMS. MMS has also been shown to effectively treat several rare cutaneous neoplasms. The future of MMS is likely to include the adoption of noninvasive imaging, immunostaining, and digital technology. CONCLUSION Mohs micrographic surgery is an effective treatment modality for numerous cutaneous neoplasms. It has achieved statistically significant superiority to surgical excision for the treatment of recurrent and high-risk NMSC. The future is likely to see increased use of noninvasive imaging, immunostaining, and digital technology.
Collapse
Affiliation(s)
- Diana K Cohen
- Skin Laser & Surgical Specialists of NY and NJ, Hackensack, New Jersey
| | - David J Goldberg
- Skin Laser & Surgical Specialists of NY and NJ, Hackensack, New Jersey
- Icahn School of Medicine at Mt. Sinai, New York, New York
| |
Collapse
|