Lesion Boundary Research Articles

Predicting Axillary Lymph Node Status With a Nomogram Based on Breast Lesion Ultrasound Features: Performance in N1 Breast Cancer Patients.

ObjectiveTo develop a nomogram for predicting axillary lymph node (ALN) metastases using the breast imaging reporting and data system (BI-RADS) ultrasound lexicon.MethodsA total of 703 patients from July 2015 to January 2018 were included in this study as a primary cohort for model construction. Moreover, 109 patients including 51 pathologically confirmed N1 patients (TNM staging) and 58 non-metastatic patients were recruited as an external validation cohort from March 2018 to August 2019. Ultrasound images and clinical information of these patients were retrospectively reviewed. The ultrasonic features based on the BI-RADS lexicon were extracted by two radiologists. The features extracted from the primary cohort were used to develop a nomogram using multivariate analysis. Internal and external validations were performed to evaluate the predictive efficacy of the nomogram.ResultsThe nomogram was based on two features (size, lesion boundary) and showed an area under the curve of 0.75 (95% confidence interval [CI], 0.70–0.79) in the primary cohort and 0.91 (95% CI, 0.84–0.97) in the external validation cohort; it achieved an 88% sensitivity in N1 patients.ConclusionThe nomogram based on BI-RADS ultrasonic features can predict breast cancer ALN status with relatively high accuracy. It has potential clinical value in improving the sensitivity and accuracy of the preoperative diagnosis of ALN metastases, especially for N1 patients.

Computational quantitative measures of Gd-EOB-DTPA enhanced MRI hepatobiliary phase images can predict microvascular invasion of small HCC

PurposeThis study was designed to preoperatively predict microvascular invasion (MVI) of solitary small hepatocellular carcinoma (sHCC) by quantitative analysis of Gd-EOB-DTPA enhanced hepatobiliary phase (HBP) magnetic resonance imaging (MRI). MethodSixty-one patients, 19 with and 42 without histologically confirmed MVI following hepatic resection for solitary sHCC (≤ 3 cm), were preoperatively examined with Gd-EOB-DTPA-enhanced MRI. The regions of interest (ROIs) of the hepatic lesions were manually delineated on the maximum cross-sectional area in the HBP images and used to calculate the lesion boundary index (LBI) and marginal gray changes (MGC). Histogram analysis was performed to measure standard deviations (STD) and coefficients of variation (CV). Correlations between quantitative parameters and MVI were evaluated and differences between MVI positive and negative groups were assessed. ResultsThe average LBI (0.85 ± 0.07) and MGC (0.48 ± 0.27) values of the negative group were significantly higher (p < 0.05) than the corresponding LBI (0.72 ± 0.07) and MGC (0.28 ± 0.18) values of the positive group. STDs and CVs in the negative group were significantly smaller (p < 0.05) than those of the positive group. Receiver operating characteristic (ROC) analysis revealed that LBI had the best predictive value with an AUC, sensitivity, and specificity of 0.91, 87 %, and 80 %, respectively. ConclusionsQuantitative analysis of HBP images is useful for predicting MVI and beneficial to clinicians in making decisions before treatment.

The application of laparoscopy combined with indocyanine green fluorescence imaging technique for hepatic cystic echinococcosis

BackgroundWith the mature application of laparoscopy in hepatobiliary surgery, laparoscopic treatment of hepatic cystic echinococcosis (CE) has made certain progress. But, due to the inherent limitations of laparoscopy and the growth characteristics of cystic echinococcosis, distinguishing the boundary between cystic lesion and normal hepatic parenchyma is pivotal importance for successful surgery. Indocyanine green (ICG) fluorescence imaging technology can view the boundary of lesion and normal tissue during the treatment of hepatic cystic echinococcosis. Applied laparoscopy combined with ICG fluorescence imaging technique for hepatic cystic echinococcosis may be an effective surgical strategy.MethodsThe clinical data contained nine patients with hepatic cystic echinococcosis who underwent laparoscopic surgery with indocyanine green fluorescence imaging technique in authors’ institution from December 2018 to December 2019 were retrospectively analyzed. Indocyanine green was administered intravenously three days prior to surgery. The fluorescence acquisition system for real-time imaging was used during the surgery and the patients were followed up after surgery.ResultsOf reported nine patients, six are male and the remaining three are female. The average age is (36.4 ± 7.6) years. For all subjects, surgical procedures were performed under laparoscopy with indocyanine green fluorescence system. This technique showed the clear boundary of the hepatic cyst with normal liver parenchyma. Total cystectomy in six patients, subtotal cystectomy in two patients and partial hepatectomy in one patient were performed respectively. The average operation time was 3.8 ± 0.9 h, blood loss 206.0 ± 120.7 ml. Neither blood transfusion nor post-operative complication was experienced. The average abdominal drainage time was 3.4 ± 0.9 days with hospital stay 5.7 ± 2.1 days. During the 6–12 months follow-up period, neither recurrence nor intraperitoneal implantation was found.ConclusionsApplied laparoscopy combined with ICG fluorescence imaging technique for hepatic cystic echinococcosis is safe and feasible. Enhanced boundary image can assist surgeons to complete radical resection and reduce complications.

Multimodal MR Features of 8 Cases of Epithelioid Glioblastoma.

Purpose The MRI features of epithelioid glioblastoma (eGBM) were analyzed. The apparent diffusion coefficient (ADC), MR perfusion-weighted imaging (PWI), and magnetic resonance spectroscopy (MRS) findings were quantitatively analyzed.Methods The MRI images of 8 cases of eGBM were analyzed retrospectively. The location and edge, signal, peritumoral edema, adjacent meningeal invasion, and enhancement of the lesions were observed. The ADC value, relative cerebral blood volume (rCBV), relative cerebral blood flow (rCBF), and N-acetylaspartate/acetylcholine (NAA/Cho) value were analyzed.Results Among the 8 patients, the tumors were mainly located in the temporal lobe (n = 3), frontal lobe (n = 3), and parietal lobe (n = 2). The lesion boundary was clear in 6 cases and unclear in 2. The lesions were superficial in 5 cases and in the deep white matter in 3. Internal hemorrhage was observed in 4 cases. There was cystic necrosis in 7 cases, and only 1 case was solid without cystic necrosis. There was no edema around the lesion in 1 case, severe edema in 5, and moderate edema in 2. In 4 cases, the adjacent meninges were involved, and in 1 case, the ependyma was involved. Two patients developed leptomeningeal metastasis within 2 months after the operation. The average ADC value of the tumor parenchyma among all 8 patients was7.15 × 10−4 mm2/s,which was 17.6% lower than that of the contralateral side. The Cho/NAA metabolite ratio was 5.27 and 0.81 in the lesions of 2 patients. The rCBV was 3.51 ml/100 g and 3.32 ml/100 g of lesions in 2 patients; these values were 36% and 29% higher, respectively, than those of the contralateral side. The rCBF was 31.5 ml/100 g/min and 82.1 ml/100 g/min of lesions in two patients; these values were 49% and 203% higher, respectively, than those of the contralateral side.Conclusion eGBM characteristics include a superficial location, easy cyst degeneration, easy necrosis and hemorrhage, and clear boundaries. It easily invades adjacent meninges and shows cerebrospinal fluid dissemination and metastasis. Combining new MR techniques, such as ADC values, PWI, and MRS, could be helpful for improving diagnostic accuracy.

Genetic algorithm-based initial contour optimization for skin lesion border detection

Automated segmentation has an essential role in detecting several diseases, such as skin lesions. In segmentation, the active contour (AC) is an efficient method based on energy forces and constraints in an image to separate the region of interest (ROI) by defining a curvature or contour. It outlines an initial contour to fit the ROI, which changed iteratively by minimizing the energy function. If the contour is improperly initialized, the AC may trap in local minima. In this work, the initial contour of the AC without edge ‘Chan-Vese’ model is optimized using the genetic algorithm (GA) to find the optimal initial circular area percentage of the skin lesion image from the whole image area. This optimal optimized value drives the AC and enhances the performance of the traditional AC while detecting the skin lesion boundaries. Various evaluation metrics were measured to compare the performance of the proposed optimized IAC (initial active contour), graph-cut, and the k-means, in dermoscopic image segmentation. The results show the dominance of the proposed method indicating that the optimal initial circular contour of 30.86% from the original image area. The results proved 96.2% detection accuracy best results achieved using this optimal value.

Advanced intraoperative ultrasound (ioUS) techniques in focal cortical dysplasia (FCD) surgery: A preliminary experience on a case series

Introduction: Focal Cortical Dysplasia (FCD) represents a broad spectrum of histopathological entities that cause drug-resistant epilepsy. Surgery has been shown to be the treatment of choice, but incomplete resection represents the leading cause of seizure persistence. Preliminary experiences with intraoperative ultrasound (ioUS) have proven its potential in defining and characterizing the lesion. In this study we analyzed the feasibility of advanced ultrasound techniques such as sono-elastography (SE) and contrast enhancement ultrasound (CEUS) in a small cohort of patients with FCD. Material and methods: We retrospectively reviewed all clinical records and images of patients with drug resistant epilepsy who underwent at least one advanced sonographic technique (SE and/or CEUS) during ioUS guided surgery between November 2014 and October 2017. We excluded from our analysis all patients with lesions other than FCD or those who had FCD associated with other pathological entities. Results: Four patients with type IIb FCD in the right frontal lobe were evaluated. All of them underwent SE, which highlighted heterogeneous stiffness in the dysplastic foci, also multiple areas of higher consistency were detected in all patients. Three patients evaluated with CEUS had visible enhancement in the FCD. Neither SE nor CEUS were better than ioUS in the identification of lesion boundaries. In the three patients who underwent both SE and CEUS we found no correspondence between stiffer areas and enhancement in the dysplastic areas. Conclusion: Ourpreliminary report confirms the feasibility of SE and CEUS in FCD surgery and describes the imaging findings in this category of patients. Studies on larger cohorts of patients are warranted to better clarify the role of these advanced intraoperative ultrasound techniques in patients with FCD.

Identification of paddy leaf diseases based on texture analysis of Blobs and color segmentation

There are three types of paddy leaf diseases that have similar symptoms, making it difficult for farmers to identify them, namely blast, brown-spot, and narrow brown-spot. This study aims to identification paddy plant diseases based on texture analysis of Blobs and color segmentation. Blobs analysis is used to get the number of objects, area and perimeter. Color segmentation is used to find out some color parameters of paddy leaf disease such as the color of the lesion boundary, the color of the spot of the lesion, and the color of the paddy leaf lesion. To get the best results, four methods have been chosen to obtained the threshold value, Otsu threshold value, variable threshold value, local threshold value and global threshold value. The best accuracy of the four methods using threshold variables is 90.7%. The results of this study indicate that the method used has been very satisfactory in identifying paddy plant disease.

Proposing a novel unsupervised stack ensemble of deep and conventional image segmentation (SEDCIS) method for localizing vitiligo lesions in skin images.

Vitiligo is an acquired pigmentary skin disorder characterized by depigmented macules and patches which brings many challenges for the patients suffering from. For vitiligo severity assessment, several scoring methods have been proposed based on morphometry and colorimetry. But, all methods suffer from much inter- and intra-observer variations for estimating the depigmented area. For all mentioned assessment methods of vitiligo disorder, accurate segmentation of the skin images for lesion detection and localization is required. The image segmentation for localizing vitiligo skin lesions has many challenges because of illumination variation, different shapes and sizes of vitiligo lesions, vague lesion boundaries and skin hairs and vignette effects. The manual image segmentation is a tedious and time-consuming task. Therefore, using automatic image segmentation methods for lesion detection is necessarily required. In this study, a novel unsupervised stack ensemble of deep and conventional image segmentation (SEDCIS) methods is proposed for localizing vitiligo lesions in skin images. Unsupervised segmentation methods do not require prior manual segmentation of vitiligo lesions which is a tedious and time-consuming task with intra- and inter-observer variations. Our collected dataset includes 877 images taken from 21 patients with the resolution of 5760*3840 pixels suffering from vitiligo disorder. Experimental results show that SEDCIS outperforms the compared methods with accuracy of 97%, sensitivity of 98%, specificity of 96%, area overlapping of 94%, and Dice index of 97%. The proposed method can segment vitiligo lesions with highly reasonable performance and can be used for assessing the vitiligo lesion surface.

Exploring Mechanisms of Quantitative Resistance to Leptosphaeria maculans (Blackleg) in the Cotyledons of Canola (Brassica napus) Based on Transcriptomic and Microscopic Analyses.

Using resistant cultivars is a common approach to managing blackleg of canola/rapeseed caused by Leptosphaeria maculans (Lm). Quantitative resistance (QR), as opposed to major-gene resistance, is of interest because it is generally more durable, due to its multi-genetic basis. However, the mechanisms and genes underlying QR are mostly unknown. In this study, potential QR modes of action in “74-44 BL” was explored. This Canadian canola cultivar showed moderate but consistent race-nonspecific resistance at the cotyledon and adult-plant stages. A susceptible cultivar, “Westar”, was used as a control. After inoculation, the lesions developed more slowly on the cotyledons of 74-44 BL than those of Westar. We used RNA sequencing (-RNA-seq) to identify genes and their functions, putatively related to this resistance, and found that genes involved in programmed cell death (PCD), reactive oxygen species (ROS), signal transduction or intracellular endomembrane transport were most differentially expressed. ROS production was assessed in relation to Lm hyphal growth and lesion size; it occurred beyond the tissue colonized by Lm in 74-44 BL and appeared to trigger rapid cell death, limiting cotyledon colonization by Lm. In contrast, Lm grew more rapidly in Westar, often catching up with the ring of ROS and surpassing lesion boundaries. It appears that QR in 74-44 BL cotyledons is associated with limited colonization by Lm possibly mediated via ROS. The RNA-seq data also showed a link between ROS, signal transduction, and endomembrane vesicle trafficking, as well as PCD in the resistance. These results provide a starting point for a better understanding of the mechanisms behind QR against Lm in canola.

Implementation and validation of an in-house geometry optimization software for SRS VMAT planning of multiple cranial metastases.

PurposeThe implementation and evaluation of an in‐house developed geometry optimization (GO) software are described. The GO script provides optimal lesion clustering, isocenter placement, and collimator angle of each arc for cranial multi‐lesion stereotactic radiosurgery (SRS) volumetric modulated arc therapy (VMAT) planning.Materials and methodsAn Eclipse‐plugin program was developed to facilitate automatic plan geometry generation for multiple metastases SRS VMAT plans. A mixed, semi‐supervised exhaustive and k‐means clustering method is used to group lesions and place isocenters. The sum of squared euclidean distance (SSED) and the boundaries of lesions’ projection from beams’ eye view are used as supervised parameters to determine the optimal isocenter numbers. The collimator angle is optimized by minimizing the sum of the MLC opening area from all gantry angles for each arc. In all, 10 clinical cases treated during 2016–2017 were compared to plan quality of GO script generated plans. Paddick gradient index (GI), conformity index (CI), and local brain volume receiving 12 Gy (local V12 Gy) around each lesion were compared.ResultFor four cases, the number of isocenters was reduced in the GO plans. For four other cases, the GO plans had the same number of isocenters as their corresponding clinical plans but with different lesion grouping. The GO plans had significantly lower GI (4.1 ± 1.0 vs 4.4 ± 0.9, P < 0.0001) and local V12 Gy (5.1 ± 4.2 vs 5.5 ± 4.3 in cm3, P < 0.0001), but not significantly different mean normal brain dose or CI. The volume of normal brain receiving ≥6 Gy was significantly lower in the GO plans. The total time to run the GO script for each case was <2 min.ConclusionThe GO software automates lesion grouping, isocenter placement, and the collimator angles for SRS VMAT planning. When tested on 10 cases, the GO script resulted in improved plan quality and shorter planning time when compared to the clinical SRS VMAT plans.

Which pathologic staining method can visualize the hyperacute infarction lesion identified by diffusion MRI?: A comparative experimental study

BackgroundThe study aimed to establish a staining method that could delineate the macroscopic lesion boundary of a hyperacute infarction depicted by diffusion-weighted MRI (DWI) and to validate the infarction boundary by comparing different staining methods. New MethodThirteen rats with 1 -h middle cerebral artery (MCA) infarction were included. Five different staining methods (Hematoxylin and eosin (H&E), Nissl, 2,3,5-triphenyltetrazolium hydrochloride (TTC), microtubule associated protein 2 (MAP2), and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) stains) were used to identify whether the hyperacute infarction could be histopathologically identified. Dice indices were compared to evaluate similarities in the lesion area ascertained by DWI and the staining methods. Through macroscopic lesion delineation, each region was subdivided into abnormal regions in all three stains (ROIA), abnormal in two stains (ROIB), and abnormal in only one (ROIC). Microscopic cellular changes were evaluated and graded according to each region. ResultsMean Dice indices of the H&E stain were significantly higher than those of the Nissl- and MAP2-stained specimens (0.83 ± 0.052, 0.58 ± 0.107, and 0.56 ± 0.059, respectively; p = 0.000). Grading scores for ROIs in the DWI abnormal lesions varied by region: ROIA exhibited the most severe damage [median (IQR), 3 (1)], followed respectively by ROIB [median (IQR), 2 (0)] and ROIC [median (IQR), 1 (0)] Comparison with existing methodsH&E stain best reflects 1 h hyperacute DWI abnormal lesions. ConclusionsH&E stain allowed for the macroscopic delineation of the 1 h DWI-abnormal lesions, while MAP2 and Nissl stains could only partially depict lesions.

Features of disturbances of microcirculation and intercellular interactions in local cold injury

Theaimof the work is to evaluate microcirculation, innervation and intercellular interactions depending on the severity and timing of cold tissue lesions.Materials and methods. 50 patients with III–IV degree of lower extremities local cold injury, as well as their blood were explored. Patients were divided into two groups depending on the period of clinical course: 5 and 30 days of cold injury. At the same time, the patients were divided into 3 groups according to the level of the lesion. The 1 group patients had only fingers frostbitten. Group 2 – patients had lesion boundary extended to the metatarsals and metatarsals. Patients with cold trauma of the proximal segments of the lower extremities belonged to group 3, had the damage level limited by the lower third of the low leg in the most cases. Studies in these groups were carried out on the 5th day after the injury. The phenomenon of lymphocytic-platelet rosette formation and the amplitude of M-response were evaluated.Results. In patients with frostbite, the increased adhesive ability of blood platelets to lymphocytes, was significantly decreased by 30 days. By the day 5, the number of lymphocyte-platelet rosettes increased 2.2 times, and the degree of adhesion 1.7 times. By the day 5 of the injury, the value of the M-response was reduced by 1.7, and the detected changes were preserved for day 30. In patients with fingers frostbitten, the percentage of LTA increased by 1.3 times, and the degree of adhesion by 1.5 times. Registered ENMG potentials decreased by 1.7 times. In group 2, the ability of platelets to adhere to lymphocytes increased by 1.6 times, and the average number of plates participated in rosette formation increased by 2.1. The registered ENMG potentials decreased by 2.5 times. In patients with the maximum level of cold tissue damage, the adhesive ability of lymphocytes and platelets increased by 3 times, and the degree of adhesion increased by 3.5 times, the amplitude of the M-response decreased by 2.6 times.Conclusions. The patients with local cold injury have been demonstrated the increasing of the platelets ability to adhesion to lymphocytes. The increasing of the number and degree of LTA was proportional to the level of tissues damaged by cold. Also, patients with frostbite had a sharp decreasing of in the amplitude of the M-response (motor response), which was getting worse with the increasing of severity of tissue damage. In the case of local cold injury, the maximum activity deviations in the activity of lymphocytic-lamellar interactions and the m-response amplitude of the electroneuromyography were recorded in the early period of cold injury, with a tendency to restore the wound process on the 30th day.

OTSC assisted EFTR for the treatment of GIST: 40 cases analysis

Background and aims To verify the safety and efficacy of over-the-scope clip (OTSC)-assisted endoscopic full-thickness resection (EFTR) for the excision of stromal tumors. Material and methods Forty patients with gastric stromal tumors treated in the Department of Gastroenterology, Binzhou Medical University Hospital from December 2015 to March 2017 were included in this study. The surgical procedures included marking the lesion boundaries, cutting open the top surface of the lesion, installing an OTS, sucking the lesion into the transparent cap of the anatomical clip which was then released, application of an endoloop for EFTR, and confirming the complete resection and pathological examination of the lesion. Statistical analysis of the tumor site and size, operation time, success rates, complications, pathological examination results, and follow-up status was performed. Results The average operation duration was 38.40 ± 24.9 min. Three cases had an incomplete resection, but the lesion was later found to have fallen off together with the OTSC. Therefore, the treatment success rate was 100%. Postoperative pathological examination revealed leiomyomas in four cases and stromal tumors in the remaining 36 cases. Conclusions OTSC-assisted EFTR is safe and effective for resection of gastrointestinal stromal tumors, especially for those <20 mm in size.

Radiotherapy of Squamous Cell Carcinoma of the Anal Canal: Search for Optimal Solutions

The increase in the number of cured patients with squamous cell anal carcinoma and the quality of life is directly related to the improvement of radiotherapy departments technical equipment in oncological clinics. Main goals are more accurate determination of the tumor volume by using modern diagnostic methods, and creating the possibility of maximum impact on the tumor and identified affected lymph nodes while limiting dose to organs at risk. During the period from 2000 to 2015, 301 patients with squamous cell anal carcinoma, stage T1-4 N 0-3 M 0-1 , were complex treated by radiation therapy with 2DRT, 3DCRT, or IMRT/VMAT in a total dose of 50-60 Gy. The use of IMRT / VMAT in comparison with conformal radiation therapy allowed to improve: immediate results (complete tumor regression) 67.5 vs 55.6 % (p = 0.071) and 3-year long-term treatment results (overall survival) 92.9 vs 88.2 % (p = 0.05), respectively, with a tendency for the reliability of local-regional control 89.9 vs 81.6 % (p = 0.179). The use of radiation therapy in its modern version, using modern imaging methods allowed to achieve high immediate and long-term oncological results with the definition of the boundaries of the tumor lesion and summing the planned dose to the primary tumor and the affected lymph nodes.

BIRADS features-oriented semi-supervised deep learning for breast ultrasound computer-aided diagnosis

We propose a novel BIRADS-SSDL network that integrates clinically-approved breast lesion characteristics (BIRADS features) into task-oriented semi-supervised deep learning (SSDL) for accurate diagnosis of ultrasound (US) images with a small training dataset. Breast US images are converted to BIRADS-oriented feature maps (BFMs) using a distance-transformation coupled with a Gaussian filter. Then, the converted BFMs are used as the input of an SSDL network, which performs unsupervised stacked convolutional auto-encoder (SCAE) image reconstruction guided by lesion classification. This integrated multi-task learning allows SCAE to extract image features with the constraints from the lesion classification task, while the lesion classification is achieved by utilizing the SCAE encoder features with a convolutional network. We trained the BIRADS-SSDL network with an alternative learning strategy by balancing the reconstruction error and classification label prediction error. To demonstrate the effectiveness of our approach, we evaluated it using two breast US image datasets. We compared the performance of the BIRADS-SSDL network with conventional SCAE and SSDL methods that use the original images as inputs, as well as with an SCAE that use BFMs as inputs. The experimental results on two breast US datasets show that BIRADS-SSDL ranked the best among the four networks, with a classification accuracy of around 94.23 ± 3.33% and 84.38 ± 3.11% on two datasets. In the case of experiments across two datasets collected from two different institutions/and US devices, the developed BIRADS-SSDL is generalizable across the different US devices and institutions without overfitting to a single dataset and achieved satisfactory results. Furthermore, we investigate the performance of the proposed method by varying the model training strategies, lesion boundary accuracy, and Gaussian filter parameters. The experimental results showed that a pre-training strategy can help to speed up model convergence during training but with no improvement of the classification accuracy on the testing dataset. The classification accuracy decreases as the segmentation accuracy decreases. The proposed BIRADS-SSDL achieves the best results among the compared methods in each case and has the capacity to deal with multiple different datasets under one model. Compared with state-of-the-art methods, BIRADS-SSDL could be promising for effective breast US computer-aided diagnosis using small datasets.

Magnetic resonance imaging/transrectal ultrasonography fusion guided seed placement in a phantom: Accuracy between 2-seed versus 1-seed strategies

PurposeTo investigate whether the 2-seed placement per Magnetic Resonance Imaging (MRI) suspicious lesion yields a higher seed placement accuracy than a 1-seed strategy on a phantom. MethodsEight olives embedded in gelatin, each simulating a prostate, underwent MRI. Three virtual spherical lesions (3, 5, and 8 mm diameters) were marked in each olive on the MRI images and co-registered to the MRI/Transrectal Ultrasonography (TRUS) fusion biopsy system. Two radiologists placed 0.5 mm fiducials, targeting the center of each virtual lesion under fusion image guidance. Half of the 8 olives in each phantom were assigned either to the 1-seed or 2-seeds per lesion strategy. Post-procedure Computed Tomography (CT) images identified each seed and were fused with MR to localize each virtual lesion and collected the seed placement error - distance between the virtual target and the corresponding seed (using the closer seed for the 2-seed strategy). Seed placement success is defined as fiducial placement within a lesion boundary. ResultsEach operator repeated the procedure on three different phantoms, and data from 209 seeds placed for 137 lesions were analyzed, with an overall error of 3.03 ± 1.52 mm. The operator skill, operator phantom procedural experience, lesion size, and number of seeds, were independently associated with the seed placement error. Seed placement success rate was higher for the 2-seed group compared to 1-seed, although the difference was not statistically significant. ConclusionsPlacing 2 seeds per MRI lesion yielded a significantly lower error compared to 1-seed strategy, although seed placement success rate was not significantly different.

652The effect of varying irrigation flow rate during irrigated radiofrequency ablation on optimising lesion shape

Abstract Funding Acknowledgements Investigator Sponsored Research grant from Boston Scientific Introduction Irrigated catheters are the standard tool for radiofrequency (RF) ablation in the left atrium and ventricles. However, pathological studies of irrigated RF lesions show a "tear-drop" shape, with the widest diameter some depth below the endocardial surface and relative endocardial sparing. This requires overlap of lesions to achieve contiguity at the endocardial surface. There has been little investigation into the effect of altering irrigation rate on lesion shape and volume. Purpose To test the hypothesis that varying the irrigation flow rate would optimise lesion shape by minimising endocardial sparing while maintaining lesion depth. Methods In an ex vivo animal heart model, irrigated ablation lesions were performed in porcine ventricular tissue at 30W using an Intellatip MiFi OI catheter with 5 different irrigation protocols: A. fixed rate (30ml/min); B. continuous reduction (30 to 2ml/min over 30s); C. continuous increase (2 to 30ml/min over 30s); D. stepwise reduction (30ml/min (10s) to 16ml/min (10s) to 2ml/min (10s)); E. stepwise increase (2ml/min (10s) to 16ml/min (10s) to 30ml/min (10s). Contact force (10g) and ablation duration (30s) were constant. Steam pops during ablation were recorded. Tissue sections were stained with triphenyltetrazolium chloride (TTC) after ablation to allow accurate measurement of lesion boundaries. Surface diameter, lesion depth, maximum diameter, and depth at maximum diameter were measured using Vernier Calipers to calculate lesion volume. Pictures of lesions were analysed further by ImageJ software to measure the degree of endocardial sparing. The optimal protocol was further tested against fixed-rate irrigation at 20W – 40W. Results 10-20 lesions were performed for each irrigation protocol. Of the 4 experimental protocols, continuous reduction (protocol B) resulted in the most optimal lesion shape (Figure). With this protocol, endocardial sparing area was significantly reduced compared to fixed-rate irrigation (1.61 vs. 2.64mm2, P &amp;lt; 0.0001), with a trend towards an increase in surface diameter (9.25 vs. 8.46mm, P = 0.08). There was no significant difference in lesion depth (5.35 vs. 4.77mm), lesion volume (374 vs. 332mm3) or maximum diameter (10.3 vs. 10.8mm). Steam pop occurred in 1 of 20 (5%) lesions in each of protocols A and B. Significantly reduced endocardial sparing with preserved volume/depth was consistent for continuous reduction when compared to fixed-rate irrigation across power settings of 20W (17ml/min), 30 W (17ml/min or 30ml/min) and 40W (30ml/min) (P &amp;lt; 0.0001). Conclusions Continuous reduction in irrigation flow rate from 30 to 2ml/min during irrigated RF ablation results in reduced endocardial sparing with preserved lesion depth and volume when compared to fixed-rate irrigation across power settings of 20 – 40W. This may allow for greater lesion spacing while maintaining endocardial contiguity and merits further investigation to improve irrigated RF ablation efficiency. Abstract Figure

Spatial Resolution and the Magnitude of Infarct Volume Measurement Error in DWI in Acute Ischemic Stroke.

Infarct volume in acute ischemic stroke is an important prognostic marker and determines endovascular treatment decisions. This study evaluates the magnitude and potential clinical impact of the error related to partial volume effects in infarct volume measurement on diffusion-weighted MR imaging in acute stroke and explores how increasing spatial resolution could reduce this error. Diffusion-weighted imaging of 393 patients with acute stroke, of whom 56 had anterior circulation large-vessel occlusion, was coregistered to standard space. Lesion boundaries were manually segmented. A 3D lesion-volume model was resampled for voxel sizes from 4 × 4 × 8 to 1 × 1 × 2 mm, and the surface-volume, corresponding to the partial volume error, was calculated. The number of cases with anterior circulation large-vessel occlusion, in which the endovascular therapy core threshold of 70 mL was contained within the margin of error, was calculated as a function of imaging resolution. The mean infarct core volume was 27.2 ± 49.9 mL. The mean surface volume was 14.7 ± 20.8 mL for 2 × 2 × 4 mm resolution and 7.4 ± 10.7 mL for 1 × 1 × 2 mm resolution. With a resolution of 2 × 2 × 4 mm, 70 mL was contained within the margin of error in 7/56 cases (12.5%) with large-vessel occlusion, while with a 1 × 1 × 2 mm voxel size, the margin of error was 3/56 (5%). The lesion-volume range of potentially misclassified lesions dropped from 46.5-94.1 mL for a 2 × 2 × 4 mm resolution to 64.4-80.1 mL for a 1 × 1 × 2 mm resolution. Partial volume effect is an important source of error in infarct volume measurement in acute stroke. Increasing spatial resolution substantially decreases the mean error. Standard use of high-resolution DWI should be considered to increase the reliability of infarct volume measurements.

Value of contrast-enhanced ultrasound in diagnosis of atypical renal space occupying lesions

Objective To evaluate the diagnostic value of contrast-enhanced ultrasound(CEUS) in renal atypical space occupying lesions. Methods The images of conventional ultrasound and CEUS in 44 patients with 47 renal atypical space occupying lesions from May 2015 to June 2019 were retrospectively analyzed, including 26 males and 18 females. The mean age of patients was(55.9±13.7)years old, ranged from 27 to 79 years old. The mean size of the lesions in conventional ultrasound was (2.5±1.3)cm, ranged from 1.0 to 6.0 cm. The location, size, echo, boundary and color flew signals of renal lesions were observed by conventional ultrasound.The degree and the pattern of enhancement, presence or absence of a clear capsule sign were analyzed by CEUS. The ultrasound diagnosis results were compared with the pathological or clinical diagnosis. Results Of the 47 lesions, 15 were malignant and 32 were benign. 19 lesions were diagnosed by operation pathology, including 13 renal cell carcinomas, 2 renal lymphomas, 3 renal cysts and 1 inflammatory granuloma; the other 28 benign tumors or pseudotumors were diagnosed by enhanced CT or MRI, and were followed up for more than 1 year, including 5 angiomyolipomas, 15 cysts, 3 renal junction cortex defects, 2 renal column hypertrophies, 1 local non-atrophic renal parenchyma, 1 hump kidney and 1 scar. Most of the renal cell carcinomas were hypoechoic (8/13), wash-in and fast (9/13), heterogeneous hypoenhancement (6/13), 9 with pseudocapsule and 6 with necrosis. 2 renal lymphomas were hypoechoic, wash-in and fast wash-out , 1 isoenhancement and 1 hypoenhancement. The angiomyolipomas were hyperechoic, wash-in and slow wash-out (4/5), 2 hypoenhancement, 2 hyperenhancement and 1 isoenhancement. Most renal cysts were anechoic(16/18). There was no enhancement in simple renal cysts. The wall or septum of complex renal cysts was thin and homogeneous enhanced, showing wash-in and slow wash-out with isoenhancement or hypoenhancement. Inflammatory granuloma was cystic-solid. The solid part was hypoenhancement. Renal column hypertrophy, hump kidney and local non-atrophic renal parenchyma were hypoechoic, showing the same enhancement pattern as renal parenchyma by CEUS. The junctional parenchyma defect and renal scar were hyperechoic, showing no enhancement. The sensitivity, specificity, accuracy and the area under the ROC curve of conventional ultrasound and CEUS in the diagnosis of benign and malignant lesions were 86.7%, 87.5%, 87.2%, 0.871 and 100.0%, 96.9%, 97.9%, 0.984.There were significant differences in the diagnosis of renal atypical space occupying lesions between two methods (P=0.03). Conclusions CEUS was better than conventional ultrasound in the diagnosis of renal atypical space occupying lesions. CEUS had advantages in the differential diagnosis of benign and malignant tumors, benign lesions and anatomical variants mimicking a renal tumor. Key words: Kidney neoplasms; Contrast-enhanced ultrasound; Diagnosis

Skin Lesion Segmentation from Dermoscopic Images Using Convolutional Neural Network.

Clinical treatment of skin lesion is primarily dependent on timely detection and delimitation of lesion boundaries for accurate cancerous region localization. Prevalence of skin cancer is on the higher side, especially that of melanoma, which is aggressive in nature due to its high metastasis rate. Therefore, timely diagnosis is critical for its treatment before the onset of malignancy. To address this problem, medical imaging is used for the analysis and segmentation of lesion boundaries from dermoscopic images. Various methods have been used, ranging from visual inspection to the textural analysis of the images. However, accuracy of these methods is low for proper clinical treatment because of the sensitivity involved in surgical procedures or drug application. This presents an opportunity to develop an automated model with good accuracy so that it may be used in a clinical setting. This paper proposes an automated method for segmenting lesion boundaries that combines two architectures, the U-Net and the ResNet, collectively called Res-Unet. Moreover, we also used image inpainting for hair removal, which improved the segmentation results significantly. We trained our model on the ISIC 2017 dataset and validated it on the ISIC 2017 test set as well as the PH2 dataset. Our proposed model attained a Jaccard Index of 0.772 on the ISIC 2017 test set and 0.854 on the PH2 dataset, which are comparable results to the current available state-of-the-art techniques.