Evaluation of an integrated spectroscopy and classification platform for point-of-care core needle biopsy assessment: Performance characteristics from ex vivo renal mass biopsies Journal Article
| Authors: | Keshavamurthy, K. N.; Dylov, D. V.; Yazdanfar, S.; Patel, D.; Silk, T.; Silk, M.; Jacques, F.; Petre, E. N.; Gonen, M.; Rekhtman, N.; Ostroverkhov, V.; Scher, H. I.; Solomon, S. B.; Durack, J. C. |
| Article Title: | Evaluation of an integrated spectroscopy and classification platform for point-of-care core needle biopsy assessment: Performance characteristics from ex vivo renal mass biopsies |
| Abstract: | Purpose: To evaluate a transmission optical spectroscopy instrument for rapid ex vivo assessment of core needle cancer biopsies (CNBs) at the point of care. Materials and Methods: CNBs from surgically resected renal tumors and nontumor regions were scanned on their sampling trays with a custom spectroscopy instrument. After extracting principal spectral components, machine learning was used to train logistic regression, support vector machines, and random decision forest (RF) classifiers on 80% of randomized and stratified data. The algorithms were evaluated on the remaining 20% of the data set held out during training. Binary classification (tumor/nontumor) was performed based on a decision threshold. Multinomial classification was also performed to differentiate between the subtypes of renal cell carcinoma (RCC) and account for potential confounding effects from fat, blood, and necrotic tissue. Classifiers were compared based on sensitivity, specificity, and positive predictive value (PPV) relative to a histopathologic standard. Results: A total of 545 CNBs from 102 patients were analyzed, yielding 5,583 spectra after outlier exclusion. At the individual spectra level, the best performing algorithm was RF with sensitivities of 96% and 92% and specificities of 90% and 89%, for the binary and multiclass analyses, respectively. At the full CNB level, RF algorithm also showed the highest sensitivity and specificity (93% and 91%, respectively). For RCC subtypes, the highest sensitivity and PPV were attained for clear cell (93.5%) and chromophobe (98.2%) subtypes, respectively. Conclusions: Ex vivo spectroscopy imaging paired with machine learning can accurately characterize renal mass CNB at the time of tissue acquisition. © 2022 SIR |
| Keywords: | adult; controlled study; human tissue; major clinical study; histopathology; area under the curve; sensitivity and specificity; quality control; tumor biopsy; renal cell carcinoma; health care quality; radical nephrectomy; kidney tumor; training; algorithm; cancer cell; needle biopsy; ex vivo study; spectroscopy; tumor diagnosis; predictive value; receiver operating characteristic; tissues; parenchyma; diagnostic test accuracy study; principal component analysis; chromophobe renal cell carcinoma; machine learning; support vector machine; optical spectroscopy; human; male; female; article; evaluation study; random forest; surgical ward; performance indicator |
| Journal Title: | Journal of Vascular and Interventional Radiology |
| Volume: | 33 |
| Issue: | 11 |
| ISSN: | 1051-0443 |
| Publisher: | Elsevier Science, Inc. |
| Date Published: | 2022-11-01 |
| Start Page: | 1408 |
| End Page: | 1415.e3 |
| Language: | English |
| DOI: | 10.1016/j.jvir.2022.07.027 |
| PUBMED: | 35940363 |
| PROVIDER: | scopus |
| PMCID: | PMC10204606 |
| DOI/URL: | |
| Notes: | Article -- Export Date: 1 November 2022 -- Source: Scopus |
Altmetric
Citation Impact
BMJ Impact Analytics
MSK Authors
Related MSK Work