TumorMap

Overview

Autonomous and Robot Guided Tumor Reconstruction and Resection via Noncontact Multi-Laser Modalities

TumorMap integrates optical coherence tomography (OCT), laser-induced endogenous fluorescence, and cuttng laser scalpel into a unified robotic platform - autonomously 3D tumor mapping and resection with submillimeter accuracy without human intervention. We demonstrate system feasibility with murine sacroma tumors for future clinical usages.

Laser I

Optical Coherence Tomography

High-resolution 3D tissue mapping for intraoperative tumor searching and boundary localization.

Laser II

Endogenous Fluorescence

Laser-induced fluorescence spectra classified by a deep learning model to distinguish tumor from healthy tissue.

Laser III

Cutting Laser Scalpel

Noncontact tumor resection via reconstructed tumor boundaries and precise robot control.

Abstract. Surgical resection of malignant solid tumors is critically dependent on the surgeon's ability to accurately identify pathological tissue and remove the tumor while preserving surrounding healthy structures. However, building an intraoperative 3D tumor model for subsequent removal faces major challenges due to the lack of high-fidelity tumor reconstruction, difficulties in developing generalized tissue models to handle the inherent complexities of tumor diagnosis, and the natural physical limitations of bimanual operation, physiologic tremor, and fatigue creep during surgery. To overcome these challenges, we introduce TumorMap, a surgical robotic platform to formulate intraoperative 3D tumor boundaries and achieve autonomous tissue resection using a set of multifunctional lasers. TumorMap integrates a three-laser mechanism (optical coherence tomography, laser-induced endogenous fluorescence, and cutting laser scalpel) combined with deep learning models to achieve fully-automated and noncontact tumor resection. We validated TumorMap in murine osteosarcoma and soft-tissue sarcoma tumor models, and established a novel histopathological workflow to estimate sensor performance. With submillimeter laser resection accuracy, we demonstrated multimodal sensor-guided autonomous tumor surgery without any human intervention.

Main Result

System overview

End-to-end demonstration of TumorMap performing autonomous tumor resection — steps of data collection, offline model training, online tissue reconstruction, tumor searching, boundary formulation to automated resection.

Movie S1 — System Demonstration

End-to-end autonomous tumor resection pipeline using the TumorMap three-laser framework on murine tumor models.

Methods

Surgical workflow

Workflow: Each stage of the TumorMap pipeline — steps of offline tumor classifier with mice tumor datasets, online tumor searching with pre-trained models, and automated tumor resection.

Movie S2 — Tumor Mapping and Resection Workflow

Offline-to-online tumor classification: Training multilayer perceptron (MLP) tumor classifiers on laser-induced fluorescence datasets (sensor referred to as "TumorID": tumor identification). The pre-trained model is then deployed online for autonomous resection.

Movie S3 — Data Collection Method

Efficient robot-guided raster scanning for collecting fluorescence spectrum data from soft tissue sarcoma (STS) and osteosarcoma (OS) murine models.

Movie S4 — Ex Vivo Experiments

Ex vivo tissue experiments for tumor (synthetic color-coded tumor) boundary estimation and resection.

Movie S5 — Mice Tumor Experiments

Validation of TumorMap's autonomous searching, mapping and resection capabilities on realistic murine osteosarcoma (OS) and soft-tissue sarcoma (STS) tumor models.

Movie S6 — Histopathology Validation

Novel histopathological workflow developed to quantitatively estimate TumorID sensor performance with dataset collection and verification.