There is an unmet clinical need for an accurate, rapid and reliable tool for margin assessment during breast-conserving surgeries. Ultrasound offers the potential for a rapid, reproducible, and non-invasive method to assess margins. However, it is challenged by certain drawbacks, including a low signal-to-noise ratio, artifacts, and the need for experience with the acquirement and interpretation of images. A possible solution might be computer-aided ultrasound evaluation. In this study, we have developed new ensemble approaches for automated breast tumor segmentation. The ensemble approaches to predict positive and close margins (distance from tumor to margin ≤ 2.0 mm) in the ultrasound images were based on 8 pre-trained deep neural networks. The best optimum ensemble approach for segmentation attained a median Dice score of 0.88 on our data set. Furthermore, utilizing the segmentation results we were able to achieve a sensitivity of 96% and a specificity of 76% for predicting a close margin when compared to histology results. The promising results demonstrate the capability of AI-based ultrasound imaging as an intraoperative surgical margin assessment tool during breast-conserving surgery.

(1) Background: Assessing the resection margins during breast-conserving surgery is an important clinical need to minimize the risk of recurrent breast cancer. However, currently there is no technique that can provide real-time feedback to aid surgeons in the margin assessment. Hyperspectral imaging has the potential to overcome this problem. To classify resection margins with this technique, a tissue discrimination model should be developed, which requires a dataset with accurate ground-truth labels. However, establishing such a dataset for resection specimens is difficult. (2) Methods: In this study, we therefore propose a novel approach based on hyperspectral unmixing to determine which pixels within hyperspectral images should be assigned to the ground-truth labels from histopathology. Subsequently, we use this hyperspectral-unmixing-based approach to develop a tissue discrimination model on the presence of tumor tissue within the resection margins of ex vivo breast lumpectomy specimens. (3) Results: In total, 372 measured locations were included on the lumpectomy resection surface of 189 patients. We achieved a sensitivity of 0.94, specificity of 0.85, accuracy of 0.87, Matthew’s correlation coefficient of 0.71, and area under the curve of 0.92. (4) Conclusion: Using this hyperspectral-unmixing-based approach, we demonstrated that the measured locations with hyperspectral imaging on the resection surface of lumpectomy specimens could be classified with excellent performance.

Breast cancer is the most frequently diagnosed type of cancer among women globally, with an estimated incidence of 2.26 million cases in the year 2020.¹ It is the leading cause of cancer death among females, with an estimated number of 685,000 deaths in the year 2020.¹ The standard of care for early-stage breast cancer is breast-conserving therapy (BCT), which involves breast-conserving surgery (BCS) followed by adjuvant radiotherapy.^2,3 In BCS, the surgeon attempts to excise the entire tumor along with a small margin of healthy tissue while providing a satisfactory cosmetic outcome. Multiple randomized clinical trials have demonstrated equivalent survival outcomes for mastectomy and BCT if clear resection margins (edges of a breast cancer specimen) are achieved during BCS.^{4 – 7} To assess whether the margins of a specimen are clear, and therefore no cancerous cells remain in the adjacent tissue, a histopathological examination is performed after surgery. If positive (tumor-involved) margins are found, the patient is usually recommended to undergo a re-excision and/or additional doses of radiotherapy,⁸ as tumor positive margins are associated with a twofold increase in ipsilateral breast tumor recurrence and an overall lower long-term survival.^{9 – 13} The required additional therapy in patients with positive margins may result in anxiety, increased clinical complications,^14,15 unsatisfactory cosmetic outcome, and^{16 – 19} increased health care costs.^{15,20 – 22} Globally, there is a lack of consensus on what constitutes a “positive margin,” and the reported rates of positive margins after BCS in the literature vary from 9% to as high as 36% for invasive breast cancer and from 4% to as high as 23% for ductal carcinoma in situ (DCIS).²³ According to Dutch guidelines, a positive margin is defined as invasive carcinoma (IC) cells reaching the inked margin over a trajectory $>4\mathrm{mm}$ or DCIS cells reaching the inked margin over any trajectory.²⁴ Focally positive margins are defined as IC cells reaching the inked margin over a trajectory $\le 4\mathrm{mm}$.²⁴ Otherwise, the margins are considered negative. For the United States, according to the guidelines of the Society of Surgical Oncology (SSO) and the American Society for Radiation Oncology (ASTRO), a margin is considered tumor-positive when ink touches the IC or when there is DCIS present within 2 mm from the resection margin.⁹

During breast-conserving surgeries, it remains challenging to accomplish adequate surgical margins. We investigated different numbers of fibers for fiber-optic diffuse reflectance spectroscopy to differentiate tumorous breast tissue from healthy tissue ex vivo up to 2 mm from the margin. Using a machine-learning classification model, the optimal performance was obtained using at least three emitting fibers (Matthew’s correlation coefficient (MCC) of 0.73), which was significantly higher compared to the performance of using a single-emitting fiber (MCC of 0.48). The percentage of correctly classified tumor locations varied from 75% to 100% depending on the tumor percentage, the tumor-margin distance and the number of fibers.

Tumor boundary identification during colorectal cancer surgery can be challenging, and incomplete tumor removal occurs in approximately 10% of the patients operated for advanced rectal cancer. In this paper, a deep learning framework for automatic tumor segmentation in colorectal ultrasound images was developed, to provide real-time guidance on resection margins using intra-operative ultrasound. A colorectal ultrasound dataset was acquired consisting of 179 images from 74 patients, with ground truth tumor annotations based on histopathology results. To address data scarcity, transfer learning techniques were used to optimize models pre-trained on breast ultrasound data for colorectal ultrasound data. A new custom gradient-based loss function (GWDice) was developed, which emphasizes the clinically relevant top margin of the tumor while training the networks. Lastly, ensemble learning methods were applied to combine tumor segmentation predictions of multiple individual models and further improve the overall tumor segmentation performance. Transfer learning outperformed training from scratch, with an average Dice coefficient over all individual networks of 0.78 compared to 0.68. The new GWDice loss function clearly decreased the average tumor margin prediction error from 1.08 mm to 0.92 mm, without compromising the segmentation of the overall tumor contour. Ensemble learning further improved the Dice coefficient to 0.84 and the tumor margin prediction error to 0.67 mm. Using transfer and ensemble learning strategies, good tumor segmentation performance was achieved despite the relatively small dataset. The developed US segmentation model may contribute to more accurate colorectal tumor resections by providing real-time intra-operative feedback on tumor margins.

This study aims to evaluate several defined specimen parameters that would allow to determine the surgical accuracy of breast-conserving surgeries (BCS) in a representative population of patients. These specimen parameters could be used to compare surgical accuracy when using novel technologies for intra-operative BCS guidance in the future. Different specimen parameters were determined among 100 BCS patients, including the ratio of specimen volume to tumor volume (resection ratio) with different optimal margin widths (0 mm, 1 mm, 2 mm, and 10 mm). Furthermore, the tumor eccentricity [maximum tumor-margin distance − minimum tumor-margin distance] and the relative tumor eccentricity [tumor eccentricity ÷ pathological tumor diameter] were determined. Different patient subgroups were compared using Wilcoxon rank sum tests. When using a surgical margin width of 0 mm, 1 mm, 2 mm, and 10 mm, on average, 19.16 (IQR 44.36), 9.94 (IQR 18.09), 6.06 (IQR 9.69) and 1.35 (IQR 1.78) times the ideal resection volume was excised, respectively. The median tumor eccentricity among the entire patient population was 11.29 mm (SD = 3.99) and the median relative tumor eccentricity was 0.66 (SD = 2.22). Resection ratios based on different optimal margin widths (0 mm, 1 mm, 2 mm, and 10 mm) and the (relative) tumor eccentricity could be valuable outcome measures to evaluate the surgical accuracy of novel technologies for intra-operative BCS guidance.