Computer-aided diagnosis for screening of lower extremity lymphedema in pelvic computed tomography images using deep learning.

Lower extremity lymphedema (LEL) is a common complication after gynecological cancer treatment, which significantly reduces the quality of life. While early diagnosis and intervention can prevent severe complications, there is currently no consensus on the optimal screening strategy for postoperative LEL. In this study, we developed a computer-aided diagnosis (CAD) software for LEL screening in pelvic computed tomography (CT) images using deep learning. A total of 431 pelvic CT scans from 154 gynecological cancer patients were used for this study. We employed ResNet-18, ResNet-34, and ResNet-50 models as the convolutional neural network (CNN) architecture. The input image for the CNN model used a single CT image at the greater trochanter level. Fat-enhanced images were created and used as input to improve classification performance. Receiver operating characteristic analysis was used to evaluate our method. The ResNet-34 model with fat-enhanced images achieved the highest area under the curve of 0.967 and an accuracy of 92.9%. Our CAD software enables LEL diagnosis from a single CT image, demonstrating the feasibility of LEL screening only on CT images after gynecologic cancer treatment. To increase the usefulness of our CAD software, we plan to validate it using external datasets.

Generation mechanisms of bowel sounds by simultaneous measurements of X-ray fluoroscopy and bowel sounds.

In clinical practice, bowel sounds are often used to assess bowel motility. However, the diagnosis differs depending on the literature because diagnoses have been based on empirically established criteria. To establish diagnostic criteria, researching the mechanism of bowel-sound occurrence is necessary. In this study, based on simultaneously measured X-ray fluoroscopy and bowel sounds, correlation and Granger causality among bowel movement, luminal content movement, and abdominal sound were estimated. The results supported our hypothesis that the bowel moves luminal contents and luminal contents generate abdominal sounds.

Influence of visceral obesity on oncologic outcome in patients with renal cell carcinoma.

OBJECTIVE: At present, computed tomography (CT) is used in almost all patients with renal tumors. We aimed to investigate the relationship between visceral adipose tissue (VAT), as assessed by CT, and various other factors in patients with renal cell carcinoma (RCC). METHODS: We undertook an examination of VAT in 117 male patients undergoing nephrectomy or partial nephrectomy at Chiba University Hospital using software designed to detect VAT in the horizontal plane of the body cavity. Pathological stage, microvascular invasion, tumor grade, performance status, C-reactive protein, BMI, hypertension, hyperlipemia, hyperglycemia, history of smoking and cause-specific survival rate were examined in relation to VAT, and multivariate Cox regression analysis was used to determine significant predictors of cause-specific survival. RESULTS: VAT in patients with stage I disease was significantly greater than that in patients with more advanced disease (p = 0.0219). VAT in patients with microvascular invasion was significantly smaller than in those without microvascular invasion (p = 0.0260). Patients with high VAT had significantly higher cumulative cause-specific survival when compared to patients with low VAT (p = 0.0257). CONCLUSION: VAT was associated with better clinical features in patients with RCC. Further study is necessary in order to clarify the role of VAT in RCC.

Objective evaluation of visibility in virtual chromoendoscopy for esophageal squamous carcinoma using a color difference formula.

Computed virtual chromoendoscopy with flexible spectral imaging color enhancement (FICE) is a new dyeless imaging technique that enhances mucosal and vascular patterns. However, a method for selecting a suitable wavelength for a particular condition has not been established. The aim of this study is to evaluate the color difference method for quality assessment of FICE images of the intrapapillary capillary loop in magnifying endoscopy for esophageal squamous cell carcinoma. The color difference between 60 microvessels and background mucosa observed using the magnifying endoscope was 8.31±2.84 SD under white light and 12.26±3.14 (p=0.0031), 11.70±4.49 (p=0.0106), and 17.49±5.40 (p<0.0001) in FICE modes A, B, and C, respectively. The visibility scores for microvessels observed by medical students were 6.00±1.12 points under white light and 11.1±2.25 (p<0.0001), 8.65±2.06 (p=0.0001), and 12.55±2.56 (p<0.0001) in FICE modes A, B, and C, respectively. Furthermore, the measurement of color difference was correlated with the visibility score assigned by medical students (Pearson's correlation coefficient=0.583, p<0.0001) In conclusion, the color difference method corresponds to human vision and is an appropriate method for evaluation of endoscopic images.

Surgical navigation using three-dimensional computed tomography images fused intraoperatively with live video.

OBJECTIVE: For novice surgeons, successful laparoscopic radical nephrectomy depends on the treatment of the renal vessels, and the use of computed tomography assists in navigation during the surgery. METHODS: To navigate during surgery, three-dimensional volume-rendered images of the renal vessels, kidneys, and surrounding organs were created intraoperatively by the surgeon using an image-processing application. A surgeon adjusted the angle of view and the size of the images according to surgical views during the operation, using a wireless mouse enclosed in a sterilized bag. These images were simultaneously fused onto live video using image-capture software. RESULTS: Novice surgeons easily constructed the augmented reality images intraoperatively and were able to successfully treat the renal vessels. CONCLUSION: The combination of three-dimensional computed tomography volume-rendered images with live video is a valuable navigation system for use intraoperatively.

Three-dimensional image processing system for the ureter and urethra using endoscopic video.

We developed new software that enables three-dimensional (3D) images of the ureter and urethra to be automatically processed from the video image of a conventional endoscope. The entire image of the ureter or urethra is displayed opened in a picture like a 3D map. The locations of lesions and the 3D structure of the prostatic urethra are depicted using this method, which enables quantitative evaluation of lesions and the obstructive potential of the prostatic urethra. Application of the technology in the urinary tract is discussed.