Periodic Reporting for period 1 - Spectra- BREAST (Multimodal Hypersprectal Imaging and Raman Spectroscopy for Intraoperative Assessment of Breast Tumor Resection Margins)
Reporting period: 2024-12-01 to 2025-11-30
Spectra-BREAST aims to revolutionize the assessment of tumor resection margins during breast conserving surgery (BCS). Currently, approximately 20% of breast cancer patients require a second operation due to incomplete removal of cancerous tissue. This results in increased costs for the healthcare system and significant negative effects on patients’ well-being. Nowadays, the standard margin assessment procedure is postoperative histopathological specimen analysis, which takes several days to complete. Although intraoperative histopathological analysis on frozen tissue is possible, it is less effective and cannot be implemented in all clinical centers. Alternative techniques have been suggested but have low diagnostic performance or take excessive time to assess the entire resected specimen. Spectra-BREAST aims to support surgeons with an intraoperative tool providing accurate and actionable feedback on the resection margin status over the whole specimen in less than 5 minutes. To this end, we will apply a radically new, high-risk multimodal approach combining two optical techniques, Hyperspectral Imaging (HSI) and fiber-optic Raman Spectroscopy (RS) with AI-based dataanalysis. Spectra-BREAST also has the potential for wider applications, going beyond breast cancer to include margin assessment for other cancers, robotic/laparoscopic surgery, and guided biopsy.
During RP1, in accordance with the Grant Agreement, the consortium focused on the technical foundations of the work (development of the necessary optical components and an initial assessment of robotic assistance), on the set-up of the infrastructure needed to enable systematic data collection and subsequent integration/validation phases, and on the set-up of the clinical study necessary for the analysis of clinical specimens. In particular, in the technological work packages (WP1–WP4), activities concentrated on early development steps that prepare the HSI and Raman subsystems and the initial collection of data for the set-up of the AI pipeline for subsequent integration work (e.g. technical specifications, modelling/simulation, dataset preparation, and architecture planning). This work will initially lead to the delivery of the HSI imaging setup, the fiber-optic Raman system, and the pre-trained AI data analysis pipeline foreseen at M18. After that, RP2 will focus on the integration of the optical and robotic systems, the analysis of clinical samples, and the development of the AI algorithms. RP3 will be mostly related to the validation of the Spectra-BREAST device.
SpectraBREAST ambitions to identify tumors on and below the resection surface (up to a depth of 2 mm), with an unprecedented high sensitivity and specificity (over 95%) in real-time.