Deep learning for image-based liver analysis - A comprehensive review focusing on malignant lesions.

Deep learning-based methods, in particular, convolutional neural networks and fully convolutional networks are now widely used in the medical image analysis domain. The scope of this review focuses on the analysis using deep learning of focal liver lesions, with a special interest in hepatocellular carcinoma and metastatic cancer; and structures like the parenchyma or the vascular system. Here, we address several neural network architectures used for analyzing the anatomical structures and lesions in the liver from various imaging modalities such as computed tomography, magnetic resonance imaging and ultrasound. Image analysis tasks like segmentation, object detection and classification for the liver, liver vessels and liver lesions are discussed. Based on the qualitative search, 91 papers were filtered out for the survey, including journal publications and conference proceedings. The papers reviewed in this work are grouped into eight categories based on the methodologies used. By comparing the evaluation metrics, hybrid models performed better for both the liver and the lesion segmentation tasks, ensemble classifiers performed better for the vessel segmentation tasks and combined approach performed better for both the lesion classification and detection tasks. The performance was measured based on the Dice score for the segmentation, and accuracy for the classification and detection tasks, which are the most commonly used metrics.

Evaluation of a novel navigation platform for laparoscopic liver surgery with organ deformation compensation using injected fiducials.

In laparoscopic liver resection, surgeons conventionally rely on anatomical landmarks detected through a laparoscope, preoperative volumetric images and laparoscopic ultrasound to compensate for the challenges of minimally invasive access. Image guidance using optical tracking and registration procedures is a promising tool, although often undermined by its inaccuracy. This study evaluates a novel surgical navigation solution that can compensate for liver deformations using an accurate and effective registration method. The proposed solution relies on a robotic C-arm to perform registration to preoperative CT/MRI image data and allows for intraoperative updates during resection using fluoroscopic images. Navigation is offered both as a 3D liver model with real-time instrument visualization, as well as an augmented reality overlay on the laparoscope camera view. Testing was conducted through a pre-clinical trial which included four porcine models. Accuracy of the navigation system was measured through two evaluation methods: liver surface fiducials reprojection and a comparison between planned and navigated resection margins. Target Registration Error with the fiducials evaluation shows that the accuracy in the vicinity of the lesion was 3.78±1.89 mm. Resection margin evaluations resulted in an overall median accuracy of 4.44 mm with a maximum error of 9.75 mm over the four subjects. The presented solution is accurate enough to be potentially clinically beneficial for surgical guidance in laparoscopic liver surgery.

Influence of sampling accuracy on augmented reality for laparoscopic image-guided surgery.

PURPOSE: This study aims to evaluate the accuracy of point-based registration (PBR) when used for augmented reality (AR) in laparoscopic liver resection surgery. MATERIAL AND METHODS: The study was conducted in three different scenarios in which the accuracy of sampling targets for PBR decreases: using an assessment phantom with machined divot holes, a patient-specific liver phantom with markers visible in computed tomography (CT) scans and in vivo, relying on the surgeon's anatomical understanding to perform annotations. Target registration error (TRE) and fiducial registration error (FRE) were computed using five randomly selected positions for image-to-patient registration. RESULTS: AR with intra-operative CT scanning showed a mean TRE of 6.9 mm for the machined phantom, 7.9 mm for the patient-specific phantom and 13.4 mm in the in vivo study. CONCLUSIONS: AR showed an increase in both TRE and FRE throughout the experimental studies, proving that AR is not robust to the sampling accuracy of the targets used to compute image-to-patient registration. Moreover, an influence of the size of the volume to be register was observed. Hence, it is advisable to reduce both errors due to annotations and the size of registration volumes, which can cause large errors in AR systems.

Mixed reality - new image technology in experimental use. / Blandet virkelighet ­ ny bildeteknologi i klinisk utprøving.

New methods for holographic visualisation provide a true three-dimensional experience of medical images. The technique is generating great interest among surgeons.

Use of mixed reality for surgery planning: Assessment and development workflow.

Meticulous preoperative planning is an important part of any surgery to achieve high levels of precision and avoid complications. Conventional medical 2D images and their corresponding three-dimensional (3D) reconstructions are the main components of an efficient planning system. However, these systems still use flat screens for visualisation of 3D information, thus losing depth information which is crucial for 3D spatial understanding. Currently, cutting-edge mixed reality systems have shown to be a worthy alternative to provide 3D information to clinicians. In this work, we describe development details of the different steps in the workflow for the clinical use of mixed reality, including results from a qualitative user evaluation and clinical use-cases in laparoscopic liver surgery and heart surgery. Our findings indicate a very high general acceptance of mixed reality devices with our applications and they were consistently rated high for device, visualisation and interaction areas in our questionnaire. Furthermore, our clinical use-cases demonstrate that the surgeons perceived the HoloLens to be useful, recommendable to other surgeons and also provided a definitive answer at a multi-disciplinary team meeting.

The effect of intraoperative imaging on surgical navigation for laparoscopic liver resection surgery.

Conventional surgical navigation systems rely on preoperative imaging to provide guidance. In laparoscopic liver surgery, insufflation of the abdomen (pneumoperitoneum) can cause deformations on the liver, introducing inaccuracies in the correspondence between the preoperative images and the intraoperative reality. This study evaluates the improvements provided by intraoperative imaging for laparoscopic liver surgical navigation, when displayed as augmented reality (AR). Significant differences were found in terms of accuracy of the AR, in favor of intraoperative imaging. In addition, results showed an effect of user-induced error: image-to-patient registration based on annotations performed by clinicians caused 33% more inaccuracy as compared to image-to-patient registration algorithms that do not depend on user annotations. Hence, to achieve accurate surgical navigation for laparoscopic liver surgery, intraoperative imaging is recommendable to compensate for deformation. Moreover, user annotation errors may lead to inaccuracies in registration processes.

Three-Dimensional Blood Vessel Segmentation and Centerline Extraction based on Two-Dimensional Cross-Section Analysis.

The segmentation of tubular tree structures like vessel systems in volumetric datasets is of vital interest for many medical applications. In this paper we present a novel, semi-automatic method for blood vessel segmentation and centerline extraction, by tracking the blood vessel tree from a user-initiated seed point to the ends of the blood vessel tree. The novelty of our method is in performing only two-dimensional cross-section analysis for segmentation of the connected blood vessels. The cross-section analysis is done by our novel single-scale or multi-scale circle enhancement filter, used at the blood vessel trunk or bifurcation, respectively. The method was validated for both synthetic and medical images. Our validation has shown that the cross-sectional centerline error for our method is below 0.8 pixels and the Dice coefficient for our segmentation is 80% ± 2.7%. On combining our method with an optional active contour post-processing, the Dice coefficient for the resulting segmentation is found to be 94% ± 2.4%. Furthermore, by restricting the image analysis to the regions of interest and converting most of the three-dimensional calculations to two-dimensional calculations, the processing was found to be more than 18 times faster than Frangi vesselness with thinning, 8 times faster than user-initiated active contour segmentation with thinning and 7 times faster than our previous method.