Vincent Agnus

This work presents the potential of hyperspectral imaging (HSI) to monitor the thermal outcome of laser ablation therapy used for minimally invasive tumor removal. Our main goal is the establishment of indicators of the thermal damage of living tissues, which can be used to assess the effect of the procedure. These indicators rely on the spectral variation of temperature-dependent tissue chromophores, i.e., oxyhemoglobin, deoxyhemoglobin, methemoglobin, and water. Laser treatment was performed at specific temperature thresholds (from 60 to 110 °C) on in-vivo animal liver and was assessed with a hyperspectral camera (500–995 nm) during and after the treatment. The indicators were extracted from the hyperspectral images after the following processing steps: the breathing motion compensation and the spectral and spatial filtering, the selection of spectral bands corresponding to specific tissue chromophores, and the analysis of the areas under the curves for each spectral band. Results show that properly combining spectral information related to deoxyhemoglobin, methemoglobin, lipids, and water allows for the segmenting of different zones of the laser-induced thermal damage. This preliminary investigation provides indicators for describing the thermal state of the liver, which can be employed in the future as clinical endpoints of the procedure outcome.

Esophagectomy often presents anastomotic leaks (AL), due to tenuous perfusion of gastric conduit fundus (GCF). Hybrid (endovascular/surgical) ischemic gastric preconditioning (IGP), might improve GCF perfusion. Sixteen pigs undergoing IGP were randomized: (1) Max-IGP (n = 6): embolization of left gastric artery (LGA), right gastric artery (RGA), left gastroepiploic artery (LGEA), and laparoscopic division (LapD) of short gastric arteries (SGA); (2) Min-IGP (n = 5): LGA-embolization, SGA-LapD; (3) Sham (n = 5): angiography, laparoscopy. At day 21 gastric tubulation occurred and GCF perfusion was assessed as: (A) Serosal-tissue-oxygenation (StO₂) by hyperspectral-imaging; (B) Serosal time-to-peak (TTP) by fluorescence-imaging; (C) Mucosal functional-capillary-density-area (FCD-A) index by confocal-laser-endomicroscopy. Local capillary lactates (LCL) were sampled. Neovascularization was assessed (histology/immunohistochemistry). Sham presented lower StO₂ and FCD-A index (41 ± 10.6%; 0.03 ± 0.03 respectively) than min-IGP (66.2 ± 10.2%, p-value = 0.004; 0.22 ± 0.02, p-value < 0.0001 respectively) and max-IGP (63.8 ± 9.4%, p-value = 0.006; 0.2 ± 0.02, p-value < 0.0001 respectively). Sham had higher LCL (9.6 ± 4.8 mL/mol) than min-IGP (4 ± 3.1, p-value = 0.04) and max-IGP (3.4 ± 1.5, p-value = 0.02). For StO₂, FCD-A, LCL, max- and min-IGP did not differ. Sham had higher TTP (24.4 ± 4.9 s) than max-IGP (10 ± 1.5 s, p-value = 0.0008) and min-IGP (14 ± 1.7 s, non-significant). Max- and min-IGP did not differ. Neovascularization was confirmed in both IGP groups. Hybrid IGP improves GCF perfusion, potentially reducing post-esophagectomy AL.

Background Clinical evaluation of the demarcation line separating ischemic from non-ischemic liver parenchyma may be challenging. Hyperspectral imaging (HSI) is a noninvasive imaging modality, which combines a camera with a spectroscope and allows quantitative imaging of tissue oxygenation. Our group developed a software to overlay HSI images onto the operative field, obtaining HSI-based enhanced reality (HYPER). The aim of the present study was to evaluate the accuracy of HYPER to identify the demarcation line after a left vascular inflow occlusion during an anatomical left hepatectomy. Materials and methods In the porcine model (n = 3), the left branches of the hepatic pedicle were ligated. Before and after vascular occlusion, HSI images based on tissue oxygenation (StO₂), obtained through the Near-Infrared index (NIR index), were regularly acquired and superimposed onto RGB video. The demarcation line was marked on the liver surface with electrocautery according to HYPER. Local lactates were measured on blood samples from the liver surface in both ischemic and perfused segments using a strip-based device. At the same areas, confocal endomicroscopy was performed. Results After ligation, HSI demonstrated a significantly lower oxygenation (NIR index) in the left medial lobe (LML) (0.27% ± 0.21) when compared to the right medial lobe (RML) (58.60% ± 12.08; p = 0.0015). Capillary lactates were significantly higher (3.07 mmol/L ± 0.84 vs. 1.33 ± 0.71 mmol/L; p = 0.0356) in the LML versus RML, respectively. Concordantly, confocal videos demonstrated the absence of blood flow in the LML and normal perfusion in the RML. Conclusions HYPER has made it possible to correctly identify the demarcation line and quantify surface liver oxygenation. HYPER could be an intraoperative tool to guide perfusion-based demarcation line assessment and segmentation.

Fluorescence-based enhanced reality (FLER) enables the quantification of fluorescence signal dynamics, which can be superimposed onto real-time laparoscopic images by using a virtual perfusion cartogram. The current practice of perfusion assessment relies on visualizing the bowel serosa. The aim of this experimental study was to quantify potential differences in mucosal and serosal perfusion levels in an ischemic colon segment.

We address the automatic segmentation of healthy and cancerous liver tissues (parenchyma, active and necrotic parts of hepatocellular carcinoma (HCC) tumor) on multiphase CT images using a deep learning approach.

INTRODUCTION: Technical difficulty and unfamiliar surgical anatomy are the main challenges in transanal total mesorectal excision. Precise 3-dimensional real-time image guidance may facilitate the safety, accuracy, and efficiency of transanal total mesorectal excision. TECHNIQUE: A preoperative CT was obtained with 10 skin fiducials and further processed to emphasize the border of the anatomical structure by 3-dimensional modeling and pelvic organ segmentation. A forced sacral tilt by placing a 10-degree wedge under the patient’s sacrum was induced to minimize pelvic organ movement caused by lithotomy position. An optical navigation system with cranial software was used. Preoperative CT images were loaded into the navigation system, and patient tracker was mounted onto the iliac bone. Once the patient-to-image paired point registration using skin fiducials was completed, the laparoscopic instrument mounted with instrument tracker was calibrated for instrument tracking. After validating the experimental setup and process of registration by navigating laparoscopic anterior resection, stereotactic navigation for transanal total mesorectal excision was performed in the low rectal neuroendocrine tumor. RESULTS: The fiducial registration error was 1.7 mm. The accuracy of target positioning was sufficient at less than 3 mm (1.8 ± 0.9 mm). Qualitative assessment using a Likert scale was well matched between the 2 observers. Of the 20 scores, 19 were judged as 4 (very good) or 5 (excellent). There was no statistical difference between mean Likert scales of the abdominal or transanal landmarks (4.4 ± 0.5 vs 4.3 ± 1.0, p = 0.965). CONCLUSIONS: Application of an existing navigation system to transanal total mesorectal excision for a low rectal tumor is feasible. The acceptable accuracy of target positioning justifies its clinical use. Further research is needed to prove the clinical need for the procedure and its impact on clinical outcomes.

Long-term morbidity after multimodal treatment for rectal cancer is suggested to be mainly made up by nerve-injury-related dysfunctions. Stereotactic navigation for rectal surgery was shown to be feasible and will be facilitated by highlighting structures at risk of iatrogenic damage. The aim of this study was to investigate the ability to make a 3D map of the pelvic nerves with magnetic resonance imaging (MRI).

Background and aims Confocal laser endomicroscopy (CLE) can provide real-time evaluation of bowel perfusion. We aimed to evaluate CLE perfusion imaging in a hemorrhagic shock model. Materials and methods Five pigs were equipped to ensure hemodynamic monitoring. Three ileostomies per animal (total n = 15) were randomly created (T0). Blood was withdrawn targeting a mean arterial pressure of 40 mmHg (shock phase, T1), for 90 min. Infusion of Ringer’s lactate was started and continued for 90 min (T2). At the different time points: (a) stomas’ mucosa was scanned with CLE; (b) capillary lactates were measured on blood obtained by puncturing stomas’ mucosa; and (c) full-thickness stomas’ biopsies were sampled for histology, mitochondrial respiratory rate (V ₀ = basal and V _ADP = respiratory rate in excess of adenosine diphosphate), and levels of superoxide anion evaluation. Functional capillary density (FCD) was measured using ad hoc software. Results Confocal scanning provided consistent and specific imaging of bowel hypoperfusion at T1: vascular hyperpermeability (blurred and enlarged capillaries) and edema (enhanced visualization of the brush border due to increased intercellular spaces and fluorescein leakage). At the end of T2, there was an improved capillary flow. FCD-A index expressed statistically significant correlation with (1) stoma capillary lactates (p = 0.023); (2) systemic capillary lactates (p = 0.031); (3) inflammation pathology score (p = 0.048); (4) central venous pressure (p = 0.0043); and (5) pulmonary artery pressure (p = 0.01). Stoma capillary lactates (mmol/L) were significantly increased at T1 (8.81 ± 4.23; p < 0.0001) and at T2 (4.77 ± 3.13; p < 0.01) when compared to T0 inclusion values (1.86 ± 0.56). V ₀ and V _ADP (pmol O₂/min/mg tissue) were both significantly decreased at T1 versus T0 (p < 0.007 and p < 0.01, respectively) and recovered by the end of reanimation (T2 vs. T0, p = n.s.). Mean O ^·−₂ production (µmol/min/mg/dry tissue) increased at T1 and significantly decreased at T2. Conclusions Confocal laser endomicroscopy (CLE) imaging could identify morphological signs congruent with biochemical markers of bowel perfusion and could be useful for assessment of stomas.