Han, Aaron C.

Background Breast phyllodes tumours (PT) are rare biphasic neoplasms consisting of epithelial and stromal components. They are classified into benign, borderline and malignant categories. Diagnosing and grading PTs present a challenge for pathologists, as there are multiple histological parameters and their own tiers. We aim to investigate the potential role of artificial intelligence (AI) as a diagnostic aid in determining PT grade. Materials and methods We investigated 15 PT whole slide images (WSIs), comprising 5 benign, 5 borderline and 5 malignant cases. We sought to classify and retrieve the most relevant WSIs by matching histological features at different patch sizes, using the Yottixel framework for WSI processing. Patches were extracted at a 20× magnification level. We then used the KimiaNet to extract feature vectors and transformed these into barcode representations. Barcodes of a query WSI were compared with those of others in the archive, allowing us to identify the most similar WSI and determine the PT grades from histological similarities. Results We utilized ‘majority‐n accuracy’ as a measure of correctness, that is when the majority of the top‐n search results have the correct diagnosis as the query patient. We achieved a maximum reported accuracy of 67%, with a 3000 × 3000 patch size when grading PT at majority voting with n = 4. Conclusion Despite the small sample size and absence of fine‐tuning, our study demonstrated the potential of AI‐based PT grade stratification using various patch sizes through histological matching. This serves as a preliminary proof of concept, with the prospect of refinement for potential routine clinical application.

e18261 Background: Multidisciplinary tumor board meetings have shown to improve quality and outcome in the care of cancer patients. Many tertiary centers conduct site specific tumor board meetings but not all the centers especially community oncology and international centers have the manpower capacity to hold site specific meetings, the overall impact of tumor boards on patients care may vary based on its structure and composition. We propose that institutional collaboration in the comprehensive care of cancer patients utilizing E-consult and E tumor board may help to improve patient care. Methods: Retrospective data of 134 patient cases seen at the American hospital Dubai between 2016-2018 was analyzed. Cases that were reviewed at the American Hospital multidisciplinary tumor board meeting (AHD-MDT) and also referred for second opinion utilizing Mayo e-consult/e-tumor board service were analyzed using electronic health record, AHD-MDT meeting minutes and reports of Mayo e-consult/e-tumor board. 3 major areas of case assessment for review were selected 1. Pathology . 2. Medical imaging. 3. Clinical recommendations. Variation in assessment and recommendations between AHD- MDT and Mayo Clinic were compared. Results: A total of 1018 cases were reviewed in the AHD-MDT between 2016 and 2018. 136 out of the 1018(13%) cases were referred for second opinion utilizing Mayo E-Consult or E-tumor board service. 117 cases were included in the analysis as there was missing data in 4 patients, 9 were duplicate and 4 were cancelled and 7 cases were not reviewed at AHD-MDT. In 78 cases pathology was reviewed at Mayo but 4 (5%) were not reviewed at AHD. In 74 (95%) cases, pathology was reviewed both at AHD and Mayo. There was change (Ch) in 2 (3%), 7(9%)updated(Ud) and no change(Ch) in 65 cases(88%). 97 cases of imaging had change(Ch) in 1 case (1%). 101 cases of e-consult/e-tumor board were assessed for clinical recommendations. There was change(Ch) in 3(3%), 2/3 cases had a change in plan due to change in the pathology. In 35(35%) the plan was updated (Ud) and 7 out of 35(20%) were due to updated(Ud) pathology. Conclusions: Our data indicates that international collaboration as part of MCCN has resulted in significant improvement in the patient care. In a select group of challenging cases, 35% had an improvement in the final treatment plan after utilization of e-consult/e-tumor board service.