Browsing by Author "Leal, Jeffrey P."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Automated lesion detection of breast cancer in [18F] FDG PET/CT using a novel AI-Based workflow
    (Frontiers, 2022-11-14) Leal, Jeffrey P.; Rowe, Steven P.; Stearns, Vered; Connolly, Roisin M.; Vaklavas, Christos; Liu, Minetta C.; Storniolo, Anna Maria; Wahl, Richard L.; Pomper, Martin G.; Solnes, Lilja B.; Medicine, School of Medicine
    Applications based on artificial intelligence (AI) and deep learning (DL) are rapidly being developed to assist in the detection and characterization of lesions on medical images. In this study, we developed and examined an image-processing workflow that incorporates both traditional image processing with AI technology and utilizes a standards-based approach for disease identification and quantitation to segment and classify tissue within a whole-body [18F]FDG PET/CT study. Methods One hundred thirty baseline PET/CT studies from two multi-institutional preoperative clinical trials in early-stage breast cancer were semi-automatically segmented using techniques based on PERCIST v1.0 thresholds and the individual segmentations classified as to tissue type by an experienced nuclear medicine physician. These classifications were then used to train a convolutional neural network (CNN) to automatically accomplish the same tasks. Results Our CNN-based workflow demonstrated Sensitivity at detecting disease (either primary lesion or lymphadenopathy) of 0.96 (95% CI [0.9, 1.0], 99% CI [0.87,1.00]), Specificity of 1.00 (95% CI [1.0,1.0], 99% CI [1.0,1.0]), DICE score of 0.94 (95% CI [0.89, 0.99], 99% CI [0.86, 1.00]), and Jaccard score of 0.89 (95% CI [0.80, 0.98], 99% CI [0.74, 1.00]). Conclusion This pilot work has demonstrated the ability of AI-based workflow using DL-CNNs to specifically identify breast cancer tissue as determined by [18F]FDG avidity in a PET/CT study. The high sensitivity and specificity of the network supports the idea that AI can be trained to recognize specific tissue signatures, both normal and disease, in molecular imaging studies using radiopharmaceuticals. Future work will explore the applicability of these techniques to other disease types and alternative radiotracers, as well as explore the accuracy of fully automated and quantitative detection and response assessment.
  • Thumbnail Image
    Item
    Updated Results of TBCRC026: Phase II Trial Correlating Standardized Uptake Value With Pathological Complete Response to Pertuzumab and Trastuzumab in Breast Cancer
    (American Society of Clinical Oncology, 2021) Connolly, Roisin M.; Leal, Jeffrey P.; Solnes, Lilja; Huang, Chiung-Yu; Carpenter, Ashley; Gaffney, Katy; Abramson, Vandana; Carey, Lisa A.; Liu, Minetta C.; Rimawi, Mothaffar; Specht, Jennifer; Storniolo, Anna Maria; Valero, Vicente; Vaklavas, Christos; Krop, Ian E.; Winer, Eric P.; Camp, Melissa; Miller, Robert S.; Wolff, Antonio C.; Cimino-Mathews, Ashley; Park, Ben H.; Wahl, Richard L.; Stearns, Vered; Medicine, School of Medicine
    Purpose: Predictive biomarkers to identify patients with human epidermal growth factor receptor 2 (HER2)-positive breast cancer who may benefit from targeted therapy alone are required. We hypothesized that early measurements of tumor maximum standardized uptake value corrected for lean body mass (SULmax) on 18F-labeled fluorodeoxyglucose positron emission tomography-computed tomography (PET-CT) would predict pathologic complete response (pCR) to pertuzumab and trastuzumab (PT). Patients and methods: Patients with stage II or III, estrogen receptor-negative, HER2-positive breast cancer received four cycles of neoadjuvant PT. 18F-labeled fluorodeoxyglucose positron emission tomography-computed tomography was performed at baseline and 15 days after PT initiation (C1D15). Eighty evaluable patients were required to test the null hypothesis that the area under the curve of percent change in SULmax by C1D15 predicting pCR is ≤ 0.65, with a one-sided type I error rate of 10%. Results: Eighty-eight women were enrolled (83 evaluable), and 85% (75 of 88) completed all four cycles of PT. pCR after PT alone was 22%. Receiver operator characteristic analysis of percent change in SULmax by C1D15 yielded an area under the curve of 0.72 (80% CI, 0.64 to 0.80; one-sided P = .12), which did not reject the null hypothesis. However, between patients who obtained pCR and who did not, a significant difference in median percent reduction in SULmax by C1D15 was observed (63.8% v 41.8%; P = .004) and SULmax reduction ≥ 40% was more prevalent (83% v 52%; P = .03; positive predictive value, 31%). Participants not obtaining a 40% reduction in SULmax by C1D15 were unlikely to obtain pCR (negative predictive value, 91%). Conclusion: Although the primary objective was not met, early changes in SULmax predict response to PT in estrogen receptor-negative and HER2-positive breast cancer. Once optimized, this quantitative imaging strategy may facilitate tailoring of therapy in this setting.