A5067396272- MRI in cancer diagnosis
- Radiomics and Machine Learning in Medical Imaging
- Breast Lesions and Carcinomas
- AI in cancer detection
- Advanced Neuroimaging Techniques and Applications
- Digital Radiography and Breast Imaging
- Medical Imaging Techniques and Applications
- Advanced MRI Techniques and Applications
- Innovation in Digital Healthcare Systems
- Breast Cancer Treatment Studies
- Human Motion and Animation
- Diverse Topics in Contemporary Research
- Cancer and Skin Lesions
- Augmented Reality Applications
- Hematological disorders and diagnostics
- Industrial Vision Systems and Defect Detection
- Endometriosis Research and Treatment
- Cancer Risks and Factors
- Vascular Malformations Diagnosis and Treatment
- Media, Gender, and Advertising
- BRCA gene mutations in cancer
- Sarcoma Diagnosis and Treatment
- Simulation and Modeling Applications
- Cardiovascular Disease and Adiposity
- Oral and Maxillofacial Pathology
Seoul St. Mary's Hospital
2015-2024
Catholic University of Korea
2015-2024
Kangnam University
2023-2024
Purpose To evaluate apparent diffusion coefficient (ADC) histogram parameters that show correlations with prognostic factors and subtypes of breast cancer. Materials Methods At 3.0T, various ADC were calculated including the entire tumor volume in 173 invasive ductal carcinomas: minimum, 10th percentile, mean, median, 90th maximum. correlated subtype. Results The mean median value was significantly higher group lymph node metastasis, HER2 positivity, a Ki‐67 <14% than negativity for...
Automated measurement and classification models with objectivity reproducibility are required for accurate evaluation of the breast cancer risk fibroglandular tissue (FGT) background parenchymal enhancement (BPE).To develop evaluate a machine-learning algorithm FGT segmentation BPE classification.Retrospective.A total 794 patients cancer, 594 assigned to development set, 200 test set.3T 1.5T; T2 -weighted, fat-saturated T1 -weighted (T1 W) dynamic contrast (DCE).Manual was performed whole...
Background Deep learning models require large‐scale training to perform confidently, but obtaining annotated datasets in medical imaging is challenging. Weak annotation has emerged as a way save time and effort. Purpose To develop deep model for 3D breast cancer segmentation dynamic contrast‐enhanced magnetic resonance (DCE‐MRI) using weak with reliable performance. Study Type Retrospective. Population Seven hundred thirty‐six women from single institution, divided into the development ( N =...
To investigate the effect of a computer-aided diagnosis (CAD) system on breast ultrasound (US) for inexperienced radiologists in describing and determining lesions.Between October 2015 to January 2017, 500 suspicious or probable benign lesions 413 patients were reviewed. Five experienced readers retrospectively reviewed each 100 according Breast Imaging Reporting Data System (BI-RADS) lexicon category, with CAD (S-detectTM). The then made final decisions by combining results their US...
The goal of this study was to perform a retrospective analysis the ultrasonographic findings associated with low-grade endometrial stromal sarcoma.Ten pathologically confirmed cases sarcoma at our institution from January 2007 April 2014 were retrospectively reviewed. All patients underwent preoperative transvaginal ultrasound. Two radiologists came consensus regarding location, size, margin, and echogenicity tumor, as well presence intratumoral cystic degeneration its extent...
Background Metastasis and multiple myeloma are common malignant bone marrow lesions which may be difficult to distinguish because of similar imaging findings. The purpose this study was determine the value adding diffusion-weighted (DWI) standard MR differentiate from metastasis. Methods 25 patients with metastasis 18 underwent 3T DWI (b = 0, 800 s/mm2) were enrolled. They all had pathologically confirmed in a treatment naïve state. Two readers who blind final diagnosis measured average ADC...
This study aimed to determine whether apparent diffusion coefficient (ADC) and morphological features on diffusion-weighted MRI (DW-MRI) can discriminate metastatic axillary lymph nodes (ALNs) from benign in patients with breast cancer. Two radiologists measured ADC, long short diameters, long-to-short diameter ratio, cortical thickness assessed eccentric thickening, loss of fatty hilum, irregular margin, asymmetry shape or number, rim sign ALNs DW-MRI categorized them into suspicious ALNs....
Background Breast cancer is a heterogeneous disease. Recent studies showed that apparent diffusion coefficient (ADC) values have various association with tumor aggressiveness and prognosis. Purpose To evaluate the value of histogram analysis ADC obtained from whole volume in invasive ductal (IDC) carcinoma situ (DCIS). Material Methods This retrospective study included 201 patients confirmed DCIS (n = 37) IDC 164). The group was divided into two groups based on presence component: IDC–DCIS...
OBJECTIVE. The objective of this study was to evaluate changes in the positive predictive value (PPV) categorization suspicious calcification for malignancy with 4th versus 5th edition BI-RADS. MATERIALS AND METHODS. A total 469 cases from 444 women (mean age, 50.1 years; age range, 23-82 years) pathologically confirmed calcifications January 2012 June 2016 were enrolled retrospective study. Two radiologists determined morphology and distribution by consensus categorized using systems...
Breast MR is the most sensitive imaging modality, but there are cases of malignant tumors that not detected in MR. This study evaluated frequency and main causes breast lesions dynamic contrast-enhanced (DCE) A total 1707 preoperative performed between 2020 2021 were included. Three radiologists individually reviewed DCE MRs found not-detected malignancy MRs. The final decided through consensus. For selected cases, images other than MRIs, such as mammography, ultrasounds, diffusion-weighted...
The purpose of this retrospective study was to investigate the association between ipsilateral recurrence ductal carcinoma in situ (DCIS) and radiomics features from DCIS contralateral normal breast on contrast enhanced MR imaging. A total 163 patients with who underwent preoperative imaging January 2010 December 2014 were included (training cohort; n = 117, validation 46). Radiomics extracted whole tumor volume early dynamic T1-subtraction images precontrast T1 images. After feature...
This study aimed to predict early breast cancer recurrence in women under 40 years of age using radiomics signature and clinicopathological information. We retrospectively investigated 155 patients with invasive who underwent MRI surgery. Through stratified random sampling, 111 were assigned as the training set, 44 validation set. Recurrence-associated factors based on within 5 during total follow-up period. A Rad-score was generated through texture analysis (3D slicer, ver. 4.8.0) least...
To maximize the user immersion and experience in a virtual space environment, visual acoustic effects must play pivotal role.However, there are opinions that quality level of visuals sound is lower than offline due to lack realism online or poor quality.It necessary provide realistic environment system accurately identify indoor outdoor within increase sense reality through interaction according movement.To this end, research should be conducted accuracy identifying space.The obtained...
Height and mammographic breast density are well-known risk factors for cancer. This study aims to investigate the association between height with cancer in a large population-based cohort of Korean women.
<title>Abstract</title> Recent advancements in genomic technologies have become critical tools for deciphering the genetic complexities of cancer tissues, enabling precision medicine strategies aimed at improving patient clinical outcomes. Here we performed a comprehensive analysis clinically annotated whole genome and transcriptome sequences from 1,364 breast cases. Our investigation provides most detailed landscapes to date, which allowed us comprehensively correlate changes with...
Automated breast ultrasonography (ABUS) has been developed in recent decades and proposed as a promising tool for overcoming the disadvantages of hand-held (HHUS) (1).Although HHUS technology progressed is still
We propose a fully automatic method to assess and improve the quality of fat saturation in breast MR images. For this purpose, three deep neural networks were trained using both actual synthetic data. Firstly, poorly saturated cases classified binary classification network. Then, regions localized segmentation Lastly, for poor cases, remaining signals retrospectively suppressed correction The results showed that our successfully identified signals.
With the rapid growth of computer vision, among them, generative artificial intelligence is being used in various industries. CartoonGAN, one generating intelligence, can create new images by changing original image into a cartoon style, which be fields such as webtoons and animations. To this end, process improving performance CartoonGAN necessary, research for should conducted. In study, to solve degradation that occurs when applied, we would like suggest direction improve applying...