Automatic Pancreatic Cyst Lesion Segmentation on EUS Images Using a Deep-Learning Approach.

The automatic segmentation of the pancreatic cyst lesion (PCL) is essential for the automated diagnosis of pancreatic cyst lesions on endoscopic ultrasonography (EUS) images. In this study, we proposed a deep-learning approach for PCL segmentation on EUS images. We employed the Attention U-Net model for automatic PCL segmentation. The Attention U-Net was compared with the Basic U-Net, Residual U-Net, and U-Net++ models. The Attention U-Net showed a better dice similarity coefficient (DSC) and intersection over union (IoU) scores than the other models on the internal test. Although the Basic U-Net showed a higher DSC and IoU scores on the external test than the Attention U-Net, there was no statistically significant difference. On the internal test of the cross-over study, the Attention U-Net showed the highest DSC and IoU scores. However, there was no significant difference between the Attention U-Net and Residual U-Net or between the Attention U-Net and U-Net++. On the external test of the cross-over study, all models showed no significant difference from each other. To the best of our knowledge, this is the first study implementing segmentation of PCL on EUS images using a deep-learning approach. Our experimental results show that a deep-learning approach can be applied successfully for PCL segmentation on EUS images.

Ultrasound elastography of the liver, spleen, and kidneys in clinically normal cats.

A large amount of overlap exists in the B-mode ultrasound appearance of normal and abnormal liver, spleen, and kidney tissues in cats. Therefore, invasive tissue sampling procedures remain the standard method for diagnosing diseases in these organs. The purpose of our study was to assess the feasibility of ultrasound elastography as a technique for improving noninvasive characterization of the feline liver, spleen, and kidneys. Elastography was performed on 10 unsedated, clinically healthy cats. Numeric (strain) values (0 = softest to 255 = firmest) assigned to color pixels within regions of interest resulted in median scores (interquartile ranges) of body wall, 207.50 (189.75-224.00); liver, 119.00 (105.00-138.25); spleen, 127.50 (121-00-142.00); right renal cortex, 83.50 (64.00-130.00); right renal near field, 125.50 (110.75-139.75); left renal cortex, 77.50 (52.00-116.25); and left renal near field, 126.00 (114.00-145.25). Strain values were not different between organs. Body wall median was the only significantly different value (P < 0.05). Strain ratio values of body wall:organ were as follows: liver, 1.76 (1.38-2.00); spleen, 1.68 (1.47-1.83); right renal cortex, 2.31 (1.61-3.15); right renal near field, 1.62 (1.41-2.01); left renal cortex, 2.66 (1.45-4.13); and left renal near field, 1.51 (1.29-1.89). Subjectively, hepatic and splenic parenchymal tissues were homogeneous in compressibility and similar in elasticity to one another. Renal cortical tissue was softer compared to medullary tissue. Findings indicated that ultrasound elastography is a feasible technique for objectively and subjectively characterizing the feline liver, spleen, and kidneys. Further research is needed in cats with confirmed diseases of these organs, to compare the diagnostic sensitivity of ultrasound elastography vs. B-mode ultrasonography.

Machine learning based prediction of recurrence after curative resection for rectal cancer.

PURPOSE: Patients with rectal cancer without distant metastases are typically treated with radical surgery. Post curative resection, several factors can affect tumor recurrence. This study aimed to analyze factors related to rectal cancer recurrence after curative resection using different machine learning techniques. METHODS: Consecutive patients who underwent curative surgery for rectal cancer between 2004 and 2018 at Gil Medical Center were included. Patients with stage IV disease, colon cancer, anal cancer, other recurrent cancer, emergency surgery, or hereditary malignancies were excluded from the study. The Synthetic Minority Oversampling Technique with Tomek link (SMOTETomek) technique was used to compensate for data imbalance between recurrent and no-recurrent groups. Four machine learning methods, logistic regression (LR), support vector machine (SVM), random forest (RF), and Extreme gradient boosting (XGBoost), were used to identify significant factors. To overfit and improve the model performance, feature importance was calculated using the permutation importance technique. RESULTS: A total of 3320 patients were included in the study. After exclusion, the total sample size of the study was 961 patients. The median follow-up period was 60.8 months (range:1.2-192.4). The recurrence rate during follow-up was 13.2% (n = 127). After applying the SMOTETomek method, the number of patients in both groups, recurrent and non-recurrent group were equalized to 667 patients. After analyzing for 16 variables, the top eight ranked variables {pathologic Tumor stage (pT), sex, concurrent chemoradiotherapy, pathologic Node stage (pN), age, postoperative chemotherapy, pathologic Tumor-Node-Metastasis stage (pTNM), and perineural invasion} were selected based on the order of permutational importance. The highest area under the curve (AUC) was for the SVM method (0.831). The sensitivity, specificity, and accuracy were found to be 0.692, 0.814, and 0.798, respectively. The lowest AUC was obtained for the XGBoost method (0.804), with a sensitivity, specificity, and accuracy of 0.308, 0.928, and 0.845, respectively. The variable with highest importance was pT as assessed through SVM, RF, and XGBoost (0.06, 0.12, and 0.13, respectively), whereas pTNM had the highest importance when assessed by LR (0.05). CONCLUSIONS: In the current study, SVM showed the best AUC, and the most influential factor across all machine learning methods except LR was found to be pT. The rectal cancer patients who have a high pT stage during postoperative follow-up are need to be more close surveillance.

A Surgical Pen-Type Probe Design for Real-Time Optical Diagnosis of Tumor Status Using 5-Aminolevulinic Acid.

A surgical microscope is large in size, which makes it impossible to be portable. The distance between the surgical microscope and the observation tissue is 15-30 cm, and the adjustment range of the right and left of the camera is a maximum of 30°. Therefore, the surgical microscope generates an attenuation (above 58%) of irradiation of the optical source owing to the long working distance (WD). Moreover, the observation of tissue is affected because of dazzling by ambient light as the optical source power is strong (55 to 160 mW/cm2). Further, observation blind spot phenomena will occur due to the limitations in adjusting the right and left of the camera. Therefore, it is difficult to clearly observe the tumor. To overcome these problems, several studies on the handheld surgical microscope have been reported. In this study, a compact pen-type probe with a portable surgical microscope is presented. The proposed surgical microscope comprises a small and portable pen-type probe that can adjust the WD between the probe and the observed tissue. In addition, it allows the adjustment of the viewing angle and fluorescence brightness. The proposed probe has no blind spots or optical density loss.

Computer-aided detection of brain metastasis on 3D MR imaging: Observer performance study.

PURPOSE: To assess the effect of computer-aided detection (CAD) of brain metastasis (BM) on radiologists' diagnostic performance in interpreting three-dimensional brain magnetic resonance (MR) imaging using follow-up imaging and consensus as the reference standard. MATERIALS AND METHODS: The institutional review board approved this retrospective study. The study cohort consisted of 110 consecutive patients with BM and 30 patients without BM. The training data set included MR images of 80 patients with 450 BM nodules. The test set included MR images of 30 patients with 134 BM nodules and 30 patients without BM. We developed a CAD system for BM detection using template-matching and K-means clustering algorithms for candidate detection and an artificial neural network for false-positive reduction. Four reviewers (two neuroradiologists and two radiology residents) interpreted the test set images before and after the use of CAD in a sequential manner. The sensitivity, false positive (FP) per case, and reading time were analyzed. A jackknife free-response receiver operating characteristic (JAFROC) method was used to determine the improvement in the diagnostic accuracy. RESULTS: The sensitivity of CAD was 87.3% with an FP per case of 302.4. CAD significantly improved the diagnostic performance of the four reviewers with a figure-of-merit (FOM) of 0.874 (without CAD) vs. 0.898 (with CAD) according to JAFROC analysis (p < 0.01). Statistically significant improvement was noted only for less-experienced reviewers (FOM without vs. with CAD, 0.834 vs. 0.877, p < 0.01). The additional time required to review the CAD results was approximately 72 sec (40% of the total review time). CONCLUSION: CAD as a second reader helps radiologists improve their diagnostic performance in the detection of BM on MR imaging, particularly for less-experienced reviewers.

A quantitative evaluation of abdominal aorta aneurysm by CT images.

An abdominal aorta aneurysm (AAA) is a disease that aortic vessel inflates abnormally. The aorta blows up continuously, which may lead to the rupture of the aorta. The mortality of rupturing the aorta is between 75 and 90% to properly treat this disease, we need to accurate measure about variation of AAA size. our team performed that AAA is reconstructed as three dimensional (3D) images by computer tomography (CT), and analyzed the elements of inflation through a geometric parameter measurement . Subjects (seven males) who undergo an AAA are enrolled for the analysis. The authors used CT images as a primary source, and obtained secondary CT images 12 months later. By means of these data, the authors constructed 3D images of AAA and performed examinations using a geometric analysis that calculates geometric parameter such as the tortuosity, diameter, saccular and so on based on volume, area of the segmented region of the CT slices that is set up by the centroids and 8 points around it. The result of the severity biomechanical factor shows increased AAA tortuosity ratio (4.9%), AAA diameter expansion ratio [cm/year] (6.8%), AAA total diameter ratio (4.7%), AAA saccular ratio (2.4%) than 12-month before. Through these results, We can plan to endovascular repair surgery to undergoing AAA patients and possible diagnosis estimation of AAA.

Usefulness of texture analysis in differentiating transient from persistent part-solid nodules(PSNs): a retrospective study.

BACKGROUND: Early discrimination between transient and persistent par-solid ground-glass nodules (PSNs) at CT is essential for patient management. The objective of our study was to retrospectively investigate the value of texture analysis in differentiating pulmonary transient and persistent PSNs in addition to clinical and CT features. METHODS: This retrospective study was performed with IRB approval and a waiver of the requirement for patients' informed consent. From January 2007 to October 2009, we identified 77 individuals (39 men and 38 women; mean age, 55 years) with 86 PSNs on thin-section chest CT. Thirty-nine PSNs in 31 individuals were transient and 47 PSNs in 46 patients were persistent. The clinical, CT, and texture features of PSNs were evaluated. To investigate the additional value of texture analysis in differentiating transient from persistent PSNs, logistic regression analysis and C-statistics were performed. RESULTS: Between transient and persistent PSNs, there were significant differences in age, gender, smoking history, and eosinophil count among the clinical features. As for thin-section CT features, there were significant differences in lesion size, solid portion size, and lesion multiplicity. In terms of texture features, there were significant differences in mean attenuation, skewness of whole PSN, attenuation ratio of whole PSN to inner solid portion, and 5-, 10-, 25-, 50-percentile CT numbers of whole PSN. Multivariate analysis revealed eosinophilia, lesion size, lesion multiplicity, mean attenuation of whole PSN, skewness of whole PSN, and 5-percentile CT number were significant independent predictors of transient PSNs. (P<0.05) C-statistics revealed that texture analysis incorporating clinical and CT features (AUC, 92.9%) showed significantly higher differentiating performance of transient from persistent PSNs compared with the clinical and CT features alone (AUC, 79.0%). (P =  0.004). CONCLUSION: Texture analysis of PSNs in addition to clinical and CT features analysis has the potential to improve the differentiation of transient from persistent PSNs.

The recent progress in quantitative medical image analysis for computer aided diagnosis systems.

Computer-aided diagnosis (CAD) has become one of the major research subjects in medical imaging and diagnostic radiology. Many different CAD schemes are being developed for use in the detection and/or characterization of various lesions found through various types of medical imaging. These imaging technologies employ conventional projection radiography, computed tomography, magnetic resonance imaging, ultrasonography, etc. In order to achieve a high performance level for a computerized diagnosis, it is important to employ effective image analysis techniques in the major steps of a CAD scheme. The main objective of this review is to attempt to introduce the diverse methods used for quantitative image analysis, and to provide a guide for clinicians.