Characterization and Evaluation of the Artemis Camera for Fluorescence-Guided Cancer Surgery

P. B. A. A. van Driel 2 M. van de Giessen 1 M. C. Boonstra 0 T. J. A. Snoeks 2 S. Keereweer 5 S. Oliveira 3 4 C. J. H. van de Velde 0 B. P. F. Lelieveldt 1 A. L. Vahrmeijer 0 C. W. G. M. Lwik 2 J. Dijkstra 1 0 Department of Surgery, Leiden University Medical Center , Leiden, The Netherlands 1 Department of Image Processing, Leiden University Medical Center , Leiden, The Netherlands 2 Department of Radiology and Molecular Imaging, Leiden University Medical Center , Leiden, The Netherlands 3 Department of Pathology, University Medical Center Utrecht , Utrecht, The Netherlands 4 Department of Biology, Utrecht University , Utrecht, The Netherlands 5 Department of Otorhinolaryngology & Head and Neck Surgery, Erasmus Medical Center , Rotterdam, The Netherlands Purpose: Near-infrared (NIR) fluorescence imaging can provide the surgeon with real-time visualization of, e.g., tumor margins and lymph nodes. We describe and evaluate the Artemis, a novel, handheld NIR fluorescence camera. Procedures: We evaluated minimal detectable cell numbers (FaDu-luc2, 7D12-IRDye 800CW), preclinical intraoperative detection of sentinel lymph nodes (SLN) using indocyanine green (ICG), and of orthotopic tongue tumors using 7D12-800CW. Results were compared with the Pearl imager. Clinically, three patients with liver metastases were imaged using ICG. Results: Minimum detectable cell counts for Artemis and Pearl were 2 105 and 4 104 cells, respectively. In vivo, seven SLNs were detected in four mice with both cameras. Orthotopic OSC-19-luc2-cGFP tongue tumors were clearly identifiable, and a minimum FaDu-luc2 tumor size of 1 mm3 could be identified. Six human malignant lesions were identified during three liver surgery procedures. Conclusions: Based on this study, the Artemis system has demonstrated its utility in fluorescence-guided cancer surgery. Introduction I as computed tomography (CT), magnetic resonance n surgery, many non-invasive imaging modalities, such imaging (MRI), single-photon emission computed tomography (SPECT), and positron emission tomography (PET), are used in a preoperative setting for the detection of tumors and for surgical planning. Translating these techniques to the operating room is challenging due to altered body positions and tissue manipulation. Therefore, the surgeon still mainly relies on visual inspection and tactile information during surgery. New intraoperative imaging modalities that support the surgeon in identifying vital structures and discriminating healthy from diseased tissues in real-time are needed, which is especially important for laparoscopic procedures where the surgeon lacks tactile information. Near-infrared (NIR) fluorescence-guided surgery (FGS) is such a novel technique [1, 2]. Compared to SPECT or PET, NIR fluorescence provides high-resolution images, can visualize microscopically tumor nodules, and can be tumorspecific due to targeted exogenous agents [3]. NIR light has the advantage of increased depth penetration and decreased autofluorescence compared to visible light [4, 5]. Furthermore, NIR light is invisible to the human eye and consequently does not alter the surgical field. The success of FGS in recognizing tumors and vital structures depends to a large extent on the imaging system used. In an excellent review, Gioux et al. [6] systematically described the required criteria to which a new clinically applicable NIR fluorescence camera system has to comply. These requirements are translated into a set of practical criteria. The most important criteria for practical application are the following: field of view, imaging distance to the patient, maneuverability, simultaneous imaging of near-infrared and visible light, real-time imaging, light intensity, sterility, and electrical safety. These criteria mainly affect the design choices of the following camera components: sensor, lens system, light source, and filters/dichroic mirrors. Currently, a small number of camera systems that fit most of the criteria above are clinically available [7]. The intraoperative Artemis imaging system is recently developed within the Center for Translational Molecular Medicine (CTMM) consortium. The system is developed in close collaboration with the clinic, which resulted in an easily maneuverable system (Fig. 1a) that acquires (NIR) fluorescence and white light images simultaneously allowing for a depicted overlay. Furthermore, the Artemis has an option to assemble a laparoscope to the camera head, allowing for minimally invasive surgery. The goal of this work was to evaluate the Artemis camera in two oncological procedures in which real-time NIR fluorescence could be of added value: (a) radical tumor resection and (b) the detection of sentinel lymph nodes, the first draining nodes from the tumor. Irradical tumor resections are a major problem in cancer surgery. At present, although tumors clinically appear to be radically resected, high percentages of microscopically irradical resection (...truncated)