Intraoperative fluorescence molecular imaging (FMI) is increasingly used to distinguish benign from malignant tumors; however, existing quantitative method

Control tissue is essential for ensuring the precision of semiquantitative analysis in back-table fluorescence imaging. However, there remains a lack of agreement on the appropriate selection of control tissues. To evaluate the back-table fluorescence imaging performance of different normal tissues and identify the optimal normal tissue, a cohort of 39 patients with orbital tumors were enrolled in the study. Prior to surgery, these patients received indocyanine green (ICG) and following resection, 43 normal control tissues (34 adipose tissues, 3 skin tissues, 3 periosteal tissues, and 3 muscle tissues) were examined using back-table fluorescence imaging. The skin tissue demonstrated significantly elevated fluorescence intensity in comparison to the diseased tissue, whereas the muscle tissue exhibited a broad range and standard deviation of fluorescence signal intensity. Conversely, the adipose and periosteum displayed weak fluorescence signals with a relatively consistent distribution. Additionally, no significant correlations were found between the signal-to-background ratio (SBR) of adipose tissue and patients’ ages, genders, weights, disease duration, tumor origins, dosing of administration of ICG infusion, and the time interval between ICG infusion and surgery. However, a positive correlation was observed between the SBR of adipose tissue and its size, with larger adipose tissues (>1 cm) showing an average SBR 27% higher than smaller adipose tissues (≤1 cm). In conclusion, the findings of this study demonstrated that adipose tissue consistently exhibited homogeneous hypofluorescence during back-table fluorescence imaging, regardless of patient clinical variables or imaging parameters. The size of the adipose tissue was identified as the primary factor influencing its fluorescence imaging characteristics, supporting its utility as an ideal control tissue for back-table fluorescence imaging.

Accurate intraoperative differentiation between malignant and benign breast tissues, particularly the assessment of lymph node status and tumor margins, is critical for surgical decision-making and prognosis. Traditional histopathological methods, such as frozen section analysis, are time-consuming and labor-intensive. Magnetic Particle Imaging (MPI) is a novel, radiation-free modality that senses the microenvironmental properties of tissues through the dynamic response of magnetic tracers. In this study, we propose a diagnostic method utilizing the higher-order harmonic response of magnetic nanoparticles. Various ex vivo breast tissue samples were immersed in Synomag-50 nanoparticles. Using a custom-built MPI spectrometer (5 kHz excitation, 9 mT amplitude) operating in spectroscopic mode, we implemented a rapid acquisition protocol in which each sample was measured 10 times, with 0.1 s per cycle. We analyzed the magnetic response spectrum and calculated the ratio of the third to the fifth harmonic ($H_3/H_5$). Histological analysis confirmed the effective infiltration of MNPs into the interstitial spaces. The repeated measurement data demonstrated high stability. A distinct stepwise increase in harmonic ratios was observed from normal tissue to tumor-adjacent tissue and finally to malignant tumors. Specifically, malignant samples showed ratios that generally exceeded 2.2, whereas benign samples remained below 2.0. These preliminary findings suggest that the harmonic ratio could serve as a sensitive biomarker reflecting the microenvironmental constraints associated with malignancy. This study validates the feasibility of utilizing MPI signal harmonics as a quantitative metric with rapid signal acquisition capabilities for differentiating benign and malignant lymph nodes.

Accurate margin assessment remains a challenge in breast-conserving surgery, where inaccuracy leads to excessive removal and compromised aesthetics. Although second near-infrared (NIR-II) fluorescence imaging offers high contrast, there is still a lack of clinically viable NIR-II equipment. We develop an innovative NIR-II fluorescence navigation platform based on indocyanine green (ICG) that enables rapid deployment, operation-friendly workflow, and precision-at-hand capability for intraoperative guidance. Thirty-seven patients were recruited with 1 mg/kg ICG injection 45 minutes before surgery. Image quality was evaluated using perception-based image quality evaluator (PIQE) score, while quantitative analyses of fluorescence, including tumor-to-normal tissue ratio (TNR) and signal-to-background ratio (SBR), together with histopathology and a multi-level reader study, were performed to assess tumor contrast and diagnostic performance. Compared to conventional NIR-I imaging, our NIR-II system demonstrates performance with enhanced spatial resolution and a higher tumor-to-normal tissue ratio (4.383 ± 1.719 versus 3.154 ± 1.072, P < 0.0001) in 37 patients. The intraoperative NIR-II imaging demonstrates superior tumor detection sensitivity [90.63% (75.78%-96.76%) vs. 84.38% (68.25%-93.14%)], with 100% diagnostic accuracy in delineating malignant tumor margins. Ex vivo NIR-II imaging enables real-time, precise visualization of tumor margins and suggested its potential to reduce the volume of excision. (NIR-II-defined volume < NIR-I-defined volume < current resection volume). Through brief training, surgeons can effectively utilize NIR-II imaging, shortening the learning curve for young surgeons in tumor and margin assessment. This study demonstrates a clinically adaptable NIR-II imaging platform that expands intraoperative visualization without changing surgical workflow or adding operational burden, supporting standardized margin assessment and improved tissue preservation in breast surgery.