Two- Versus 8-Zone Lung Ultrasound in Heart Failure: Analysis of a Large Data Set Using a Deep Learning Algorithm.

OBJECTIVE: Scanning protocols for lung ultrasound often include 8 or more lung zones, which may limit real-world clinical use. We sought to compare a 2-zone, anterior-superior thoracic ultrasound protocol for B-line artifact detection with an 8-zone approach in patients with known or suspected heart failure using a deep learning (DL) algorithm. METHODS: Adult patients with suspected heart failure and B-lines on initial lung ultrasound were enrolled in a prospective observational study. Subjects received daily ultrasounds with a hand-held ultrasound system using an 8-zone protocol (right and left anterior/lateral and superior/inferior). A previously published deep learning algorithm that rates severity of B-lines on a 0-4 scale was adapted for use on hand-held ultrasound full video loops. Average severities for 8 and 2 zones were calculated utilizing DL ratings. Bland-Altman plot analyses were used to assess agreement and identify bias between 2- and 8-zone scores for both primary (all patients, 5728 videos, 205 subjects) and subgroup (confirmed diagnosis of heart failure or pulmonary edema, 4464 videos, 147 subjects) analyses. RESULTS: Bland-Altman plot analyses revealed excellent agreement for both primary and subgroup analyses. The absolute difference on the 4-point scale between 8- and 2-zone average scores was not significant for the primary dataset (0.03; 95% CI -0.01 to 0.07) or the subgroup (0.01; 95% CI -0.04 to 0.06). CONCLUSION: Utilization of a 2-zone, anterior-superior thoracic ultrasound protocol provided similar severity information to an 8-zone approach for a dataset of subjects with known or suspected heart failure.

Automated Lung Ultrasound B-Line Assessment Using a Deep Learning Algorithm.

Shortness of breath is a major reason that patients present to the emergency department (ED) and point-of-care ultrasound (POCUS) has been shown to aid in diagnosis, particularly through evaluation for artifacts known as B-lines. B-line identification and quantification can be a challenging skill for novice ultrasound users, and experienced users could benefit from a more objective measure of quantification. We sought to develop and test a deep learning (DL) algorithm to quantify the assessment of B-lines in lung ultrasound. We utilized ultrasound clips ( n = 400 ) from an existing database of ED patients to provide training and test sets to develop and test the DL algorithm based on deep convolutional neural networks. Interpretations of the images by algorithm were compared to expert human interpretations on binary and severity (a scale of 0-4) classifications. Our model yielded a sensitivity of 93% (95% confidence interval (CI) 81%-98%) and a specificity of 96% (95% CI 84%-99%) for the presence or absence of B-lines compared to expert read, with a kappa of 0.88 (95% CI 0.79-0.97). Model to expert agreement for severity classification yielded a weighted kappa of 0.65 (95% CI 0.56-074). Overall, the DL algorithm performed well and could be integrated into an ultrasound system in order to help diagnose and track B-line severity. The algorithm is better at distinguishing the presence from the absence of B-lines but can also be successfully used to distinguish between B-line severity. Such methods could decrease variability and provide a standardized method for improved diagnosis and outcome.

Design and control of an image-guided robot for spine surgery in a hybrid OR.

BACKGROUND: Minimally invasive spine (MIS) fusion surgery requires image guidance and expert manual dexterity for a successful, efficient, and accurate pedicle screw placement. Operating room (OR)-integrated robotic solution can provide precise assistance to potentially minimize complication rates and facilitate difficult MIS procedures. METHODS: A 5-degrees of freedom robot was designed specifically for a hybrid OR with integrated surgical navigation for guiding pedicle screw pilot holes. The system automatically aligns an instrument following the surgical plan using only instrument tracking feedback. Contrary to commercially available robotic systems, no tracking markers on the robotic arm are required. The system was evaluated in a cadaver study. RESULTS: The mean targeting error (N = 34) was 1.27±0.57 mm and 1.62±0.85°, with 100% of insertions graded as clinically acceptable. CONCLUSIONS: A fully integrated robotic guidance system, including intra-op imaging, planning, and physical guidance with optimized robot design and control, can improve workflow and provide pedicle screw guidance with less than 2 mm targeting error.

Clinical application of high frequency jet ventilation in stereotactic liver ablations - a methodological study.

Background: Computer-assisted navigation during thermal ablation of liver tumours, may help to correct needle placement and improve ablation efficacy in percutaneous, laparoscopic and open interventions. The potential advantage of using high frequency jet-ventilation technique (HFJV) during the procedure is by minimising the amplitude of respiration-related upper-abdominal organs movements. The aim of this clinical methodological trial was to establish whether HFJV would give less ventilatory induced liver movements than conventional ventilation, during stereotactic navigated ablation of liver metastases under open surgery. Methods: Five consecutive patients scheduled for elective, open liver ablation under general propofol and remifentanil anaesthesia were included in the study protocol. During the stereotactic targeting of the tumours, HFJV was chosen for intraoperative lung ventilation. For tracking of liver movement, a rigid marker shield was placed on the liver surface and tracked with an optical position measurement system. A 4D position of the marker shield was measured for HFJV and conventional tidal volume lung ventilation (TV). At each time point the magnitude of liver displacement was calculated as an Euclidean distance between translational component of the marker shield's 3D position and previously estimated centroid of the translational motion. Results: The mean Euclidean liver displacement was 0.80 (0.10) mm for HFJV and 2,90 (1.03) mm for TV with maximum displacement going as far as 12 mm on standard ventilation (p=0.0001). Conclusion: HFJV is a valuable lung ventilation method for patients undergoing stereotactic surgical procedures in general anaesthesia when reduction of organ displacement is crucial.

Stereotactic CT-Guided Percutaneous Microwave Ablation of Liver Tumors With the Use of High-Frequency Jet Ventilation: An Accuracy and Procedural Safety Study.

OBJECTIVE: The purpose of the present study is to evaluate the accuracy and safety of antenna placement performed with the use of a CT-guided stereotactic navigation system for percutaneous ablation of liver tumors and to assess the safety of high-frequency jet ventilation for target motion control. MATERIALS AND METHODS: Twenty consecutive patients with malignant liver lesions for which surgical resection was contraindicated or that were not readily visible on ultrasound or not accessible by ultrasound guidance were included in the study. Patients were treated with percutaneous microwave ablation performed using a CT-guided stereotactic navigation system. High-frequency jet ventilation was used to reduce liver motion during all interventions. The accuracy of antenna placement, the number of needle readjustments required, overall safety, and the radiation doses were assessed. RESULTS: Microwave ablation was completed for 20 patients (28 lesions). Performance data could be evaluated for 17 patients with 25 lesions (mean [± SD] lesion diameter, 14.9 ± 5.9 mm; mean lesion location depth, 87.5 ± 27.3 mm). The antennae were placed with a mean lateral error of 4.0 ± 2.5 mm, a depth error of 3.4 ± 3.2 mm, and a total error of 5.8 ± 3.2 mm in relation to the intended target. The median number of antenna readjustments required was zero (range, 0-1 adjustment). No major complications were related to either the procedure or the use of high-frequency jet ventilation. The mean total patient radiation dose was 957.5 ± 556.5 mGy × cm, but medical personnel were not exposed to irradiation. CONCLUSION: Percutaneous microwave ablation performed with CT-guided stereotactic navigation provides sufficient accuracy and requires almost no repositioning of the needle. Therefore, it is technically feasible and applicable for safe treatments.

A robotic needle-positioning and guidance system for CT-guided puncture: Ex vivo results.

PURPOSE: To test the feasibility of a robotic needle-guidance platform during CT-guided puncture ex vivo. MATERIAL AND METHODS: Thin copper wires inserted into a torso phantom served as targets. The phantom was placed on a carbon plate and the robot-positioning unit (RPU) of the guidance platform (iSYS Medizintechnik GmbH, Kitzbuehel, Austria) was attached. Following CT imaging and automatic registration a double oblique trajectory was planned and the RPU was remotely moved into appropriate position and angulation. A 17G-puncture needle was then manually inserted until the preplanned depth, permanently guided by the RPU. The CT scan was repeated and the distance between the actual needle tip and the target was evaluated. RESULTS: Automatic registration was successful in ten experiments and the median duration of an experiment was 9.6 (6.4-46.0) minutes. The angulation of the needle path in x-y and z-axis was within 15.6° to 32.6°, and -32.8° to 3.2°, respectively and the needle insertion depth was 92.8 ± 14.4 mm. The Euclidean distance between the actual needle tip and the target was 2.3 ± 0.8 (range, 0.9-3.7) mm. CONCLUSION: Automatic registration and accurate needle placement close to small targets was demonstrated. Study settings and torso phantom were very close to the clinical reality.

Angiographic C-arm CT- versus MDCT-guided stereotactic punctures of liver lesions: nonrigid phantom study.

OBJECTIVE: Angiographic C-arm CT may allow performing percutaneous stereotactic tumor ablations in the interventional radiology suite. Our purpose was to evaluate the accuracy of using C-arm CT for single and multimodality image fusions and to compare the targeting accuracy of liver lesions with the reference standard of MDCT. MATERIALS AND METHODS: C-arm CT and MDCT scans were obtained of a nonrigid rapid prototyping liver phantom containing five 1-mm targets that were placed under skin-simulating deformable plastic foam. Target registration errors of image fusion were evaluated for single-modality and multimodality image fusions. A navigation system and stereotactic aiming device were used to evaluate target positioning errors on postinterventional scans with the needles in place fused with the C-arm CT or MDCT planning images. RESULTS: Target registration error of the image fusion showed no significant difference (p > 0.05) between both modalities. In five series with a total of 25 punctures for each modality, the lateral target positioning error (i.e., the lateral distance between the needle tip and the planned trajectory) was similar for C-arm CT (mean [± SD], 1.6 ± 0.6 mm) and MDCT (1.82 ± .97 mm) (p = 0.33). CONCLUSION: In a nonrigid liver phantom, angiographic C-arm CT may provide similar image fusion accuracy for comparison of intra- and postprocedure control images with the planning images and enables stereotactic targeting accuracy similar to that of MDCT.

Cone-beam computed tomography-guided stereotactic liver punctures: a phantom study.

PURPOSE: Images from computed tomography (CT), combined with navigation systems, improve the outcomes of local thermal therapies that are dependent on accurate probe placement. Although the usage of CT is desired, its availability for time-consuming radiological interventions is limited. Alternatively, three-dimensional images from C-arm cone-beam CT (CBCT) can be used. The goal of this study was to evaluate the accuracy of navigated CBCT-guided needle punctures, controlled with CT scans. METHODS: Five series of five navigated punctures were performed on a nonrigid phantom using a liver specific navigation system and CBCT volumetric dataset for planning and navigation. To mimic targets, five titanium screws were fixed to the phantom. Target positioning accuracy (TPECBCT) was computed from control CT scans and divided into lateral and longitudinal components. Additionally, CBCT-CT guidance accuracy was deducted by performing CBCT-to-CT image coregistration and measuring TPECBCT-CT from fused datasets. Image coregistration was evaluated using fiducial registration error (FRECBCT-CT) and target registration error (TRECBCT-CT). RESULTS: Positioning accuracies in lateral directions pertaining to CBCT (TPECBCT = 2.1 ± 1.0 mm) were found to be better to those achieved from previous study using CT (TPECT = 2.3 ± 1.3 mm). Image coregistration error was 0.3 ± 0.1 mm, resulting in an average TRE of 2.1 ± 0.7 mm (N = 5 targets) and average Euclidean TPECBCT-CT of 3.1 ± 1.3 mm. CONCLUSIONS: Stereotactic needle punctures might be planned and performed on volumetric CBCT images and controlled with multidetector CT with positioning accuracy higher or similar to those performed using CT scanners.