Artificial Intelligence for Breast Cancer Screening in Mammography (AI-STREAM): A Prospective Multicenter Study Design in Korea Using AI-Based CADe/x / 한국유방암학회지

Purpose@#Artificial intelligence (AI)-based computer-aided detection/diagnosis (CADe/x) has helped improve radiologists’ performance and provides results equivalent or superior to those of radiologists’ alone. This prospective multicenter cohort study aims to generate real-world evidence on the overall benefits and disadvantages of using AI-based CADe/x for breast cancer detection in a population-based breast cancer screening program comprising Korean women aged ≥ 40 years. The purpose of this report is to compare the diagnostic accuracy of radiologists with and without the use of AI-based CADe/x in mammography readings for breast cancer screening of Korean women with average breast cancer risk. @*Methods@#Approximately 32,714 participants will be enrolled between February 2021 and December 2022 at 5 study sites in Korea. A radiologist specializing in breast imaging will interpret the mammography readings with or without the use of AI-based CADe/x. If recall is required, further diagnostic workup will be conducted to confirm the cancer detected on screening. The findings will be recorded for all participants regardless of their screening status to identify study participants with breast cancer diagnosis within both 1 year and 2 years of screening. The national cancer registry database will be reviewed in 2026 and 2027, and the results of this study are expected to be published in 2027. In addition, the diagnostic accuracy of general radiologists and radiologists specializing in breast imaging from another hospital with or without the use of AI-based CADe/x will be compared considering mammography readings for breast cancer screening.DiscussionThe Artificial Intelligence for Breast Cancer Screening in Mammography (AI-STREAM) study is a prospective multicenter study that aims to compare the diagnostic accuracy of radiologists with and without the use of AI-based CADe/x in mammography readings for breast cancer screening of women with average breast cancer risk. AI-STREAM is currently in the patient enrollment phase.Trial RegistrationClinicalTrials.gov Identifier: NCT05024591

Improving the Performance of Radiologists Using Artificial Intelligence-Based Detection Support Software for Mammography: A Multi-Reader Study

Objective@#To evaluate whether artificial intelligence (AI) for detecting breast cancer on mammography can improve the performance and time efficiency of radiologists reading mammograms. @*Materials and Methods@#A commercial deep learning-based software for mammography was validated using external data collected from 200 patients, 100 each with and without breast cancer (40 with benign lesions and 60 without lesions) from one hospital. Ten readers, including five breast specialist radiologists (BSRs) and five general radiologists (GRs), assessed all mammography images using a seven-point scale to rate the likelihood of malignancy in two sessions, with and without the aid of the AI-based software, and the reading time was automatically recorded using a web-based reporting system. Two reading sessions were conducted with a two-month washout period in between. Differences in the area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, and reading time between reading with and without AI were analyzed, accounting for data clustering by readers when indicated. @*Results@#The AUROC of the AI alone, BSR (average across five readers), and GR (average across five readers) groups was 0.915 (95% confidence interval, 0.876–0.954), 0.813 (0.756–0.870), and 0.684 (0.616–0.752), respectively. With AI assistance, the AUROC significantly increased to 0.884 (0.840–0.928) and 0.833 (0.779–0.887) in the BSR and GR groups, respectively (p = 0.007 and p < 0.001, respectively). Sensitivity was improved by AI assistance in both groups (74.6% vs. 88.6% in BSR, p < 0.001; 52.1% vs. 79.4% in GR, p < 0.001), but the specificity did not differ significantly (66.6% vs. 66.4% in BSR, p = 0.238; 70.8% vs. 70.0% in GR, p = 0.689). The average reading time pooled across readers was significantly decreased by AI assistance for BSRs (82.73 vs. 73.04 seconds, p < 0.001) but increased in GRs (35.44 vs. 42.52 seconds, p < 0.001). @*Conclusion@#AI-based software improved the performance of radiologists regardless of their experience and affected the reading time.

Ultrasonographic features of pure ductal carcinoma in situ of the breast: correlations with pathologic features and biological markers

PURPOSE: The purpose of this study was to evaluate the ultrasonographic features of pure ductal carcinoma in situ (DCIS) of the breast and to evaluate the correlations of ultrasonographic features with pathologic and biological features. METHODS: A total of 141 lesions in 138 women with pure DCIS who underwent preoperative breast ultrasonography were retrospectively reviewed. Ultrasonographic features were analyzed using the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) ultrasonography lexicon and the diagnostic criteria of the Japan Society of Ultrasonics in Medicine. Pathologic features including the nuclear grade and presence of comedonecrosis were evaluated. Biological markers including estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 (HER2) status, as well as the Ki-67 index, were recorded. Ultrasonographic features were compared with pathologic findings and biological markers using the chi-square test. P-values of < 0.05 were considered to indicate statistical significance. RESULTS: Of the 141 lesions, 75 (53.2%) were mass lesions, 56 (39.7%) were non-mass lesions, and 10 (7.1%) were not visible. The most common feature of the mass pattern was a mass with irregular shape (32.6%), an indistinct margin (27.7%), and hypoechogenicity (37.6%). Microcalcifications were observed in 48 cases (36.6%) as an associated feature. Calcifications outside of a mass were more common than calcifications within a mass. Ultrasonographic microcalcifications and ductal changes were frequently observed in non-mass lesions. Ultrasonographic non-mass lesions were associated with high-grade DCIS (P=0.004) and the presence of comedonecrosis (P=0.006). Microcalcifications were significantly associated with high-grade DCIS (P < 0.001), the presence of comedonecrosis (P < 0.001), an elevated Ki-67 (P < 0.001), and HER2 positivity (P=0.003). CONCLUSION: The most common ultrasonographic feature of pure DCIS was an irregular, hypoechoic mass with an indistinct margin. Ultrasonographic microcalcifications and ductal changes were more frequent in non-mass lesions, which were correlated with poor prognostic factors, such as a high nuclear grade, comedonecrosis, HER2 positivity, and an elevated Ki-67 index.

Estimation of T2* Relaxation Time of Breast Cancer: Correlation with Clinical, Imaging and Pathological Features

OBJECTIVE: The purpose of this study was to estimate the T2* relaxation time in breast cancer, and to evaluate the association between the T2* value with clinical-imaging-pathological features of breast cancer. MATERIALS AND METHODS: Between January 2011 and July 2013, 107 consecutive women with 107 breast cancers underwent multi-echo T2*-weighted imaging on a 3T clinical magnetic resonance imaging system. The Student's t test and one-way analysis of variance were used to compare the T2* values of cancer for different groups, based on the clinical-imaging-pathological features. In addition, multiple linear regression analysis was performed to find independent predictive factors associated with the T2* values. RESULTS: Of the 107 breast cancers, 92 were invasive and 15 were ductal carcinoma in situ (DCIS). The mean T2* value of invasive cancers was significantly longer than that of DCIS (p = 0.029). Signal intensity on T2-weighted imaging (T2WI) and histologic grade of invasive breast cancers showed significant correlation with T2* relaxation time in univariate and multivariate analysis. Breast cancer groups with higher signal intensity on T2WI showed longer T2* relaxation time (p = 0.005). Cancer groups with higher histologic grade showed longer T2* relaxation time (p = 0.017). CONCLUSION: The T2* value is significantly longer in invasive cancer than in DCIS. In invasive cancers, T2* relaxation time is significantly longer in higher histologic grades and high signal intensity on T2WI. Based on these preliminary data, quantitative T2* mapping has the potential to be useful in the characterization of breast cancer.

Estimation of T2* Relaxation Times for the Glandular Tissue and Fat of Breast at 3T MRI System

PURPOSE: T2* relaxation time which includes susceptibility information represents unique feature of tissue. The objective of this study was to investigate T2* relaxation times of the normal glandular tissue and fat of breast using a 3T MRI system. MATERIALS AND METHODS: Seven-echo MR Images were acquired from 52 female subjects (age 49 +/- 12 years; range, 25 to 75) using a three-dimensional (3D) gradient-echo sequence. Echo times were between 2.28 ms to 25.72 ms in 3.91 ms steps. Voxel-based T2* relaxation times and R2* relaxation rate maps were calculated by using the linear curve fitting for each subject. The 3D regions-of-interest (ROI) of the normal glandular tissue and fat were drawn on the longest echo-time image to obtain T2* and R2* values. Mean values of those parameters were calculated over all subjects. RESULTS: The 3D ROI sizes were 4818 +/- 4679 voxels and 1455 +/- 785 voxels for the normal glandular tissue and fat, respectively. The mean T2* values were 22.40 +/- 5.61 ms and 36.36 +/- 8.77 ms for normal glandular tissue and fat, respectively. The mean R2* values were 0.0524 +/- 0.0134/ms and 0.0297 +/- 0.0069/ms for the normal glandular tissue and fat, respectively. CONCLUSION: T2* and R2* values were measured from human breast tissues. T2* of the normal glandular tissue was shorter than that of fat. Measurement of T2* relaxation time could be important to understand susceptibility effects in the breast cancer and the normal tissue.

Investigation of Correlations of Double Inversion Recovery and MR Spectroscopy on Breast MR Imaging

PURPOSE: To evaluate the correlation of lesion-to-normal ratio (LNR) of signal intensity from double inversion recovery MR imaging and total choline-containing compound (tCho) resonance from single voxel MR spectroscopy in breast cancers. MATERIALS AND METHODS: Between August 2008 and December 2009, 28 patients who were diagnosed as breast cancer and had undergone both double inversion recovery (DIR) MR imaging and MR spectroscopy (MRS) were included in this study. The signal intensities of the lesion (L) and ipsilateral normal breast tissue (N) were measured in region of interest of each breast cancer in DIR and contrast enhance MR image (CE-T1WI) to calculate the LNR value for each technique. MRS was performed using single-voxel MR spectroscopy. The height, width and area of tCho resonance were compared with each LNR of DIR and CE-T1WI. We used Pearson's correlation coefficient (r) for correlation analysis and the significance level was p=0.05. RESULTS: There was no statistically significant correlation between LNR of CE-T1WI and height (r=-0.322, p=0.094), width (r=-0.233, p=0.232) and area (r=-0.309, p=0.109) of MRS tCho. There was no statistically significant correlation between LNR of DIR and height (r=0.067, p=0.735), width (r=-0.287, p=0.139) and area (r=0.012, p=0.953) of MRS tCho, either. The Pearson's correlation coefficient was 0.186 between LNRs of CE-T1WI and DIR (p=0.344). CONCLUSION: There was no statistically significant correlation between LNR of DIR and relative amount of tCho resonance of MRS.

Multiple Symmetric Lipomatosis (Madelung's Disease) Presenting as Bilateral Huge Gynecomastia / 한국유방암학회지

Multiple symmetric lipomatosis (MSL), or Madelung's disease, is a rare disease of unknown etiology. It is characterized by the presence of loose adipose tissue deposits localized in the cervical region and upper body. MSL presenting as bilateral huge gynecomastia is an extremely rare phenomenon. The present report describes a case of MSL in a 66-year-old man. The patients presented with bilateral breast bulging. He had a history of cigarette and alcohol use. His condition was treated with a bilateral nipple-sparing mastectomy. MSL can present as a form of gynecomastia, for its accurate diagnosis and proper treatment of MSL, increasing awareness of the clinical characteristics of the disease is required, especially amongst breast surgeons. Herein, we review the literature and discuss the clinical characteristics, pathology, and surgical treatment of MSL.

BRAF mutation may predict higher necessity of postoperative radioactive iodine ablation in papillary thyroid cancer

PURPOSE: The primary aim of the present study was to analyze the association between high-risk clinicopathologic characteristics and the BRAFV600E mutation. METHODS: From March 2010 to September 2012, we performed analysis of the BRAF mutation (assessing V600E point mutation of BRAF gene, exon 15, on chromosome 7q34 by real-time polymerase chain reaction kit) from 499 papillary thyroid carcinoma (PTC) patients who underwent thyroidectomy. We analyzed the relation between the mutation and known clinicopathologic risk factors of PTC. RESULTS: BRAF mutations were found in 353 of 499 patients (70.7%). On univariate analysis, BRAF mutations were more frequently detected in patients with central lymph node metastasis (78.5% vs. 66.7%, P = 0.007) and classic PTC type (71.3% vs. 16.7%, P = 0.011). Patients with one or more aggressive pathologic feature such as lymph node metastasis, multifocality, and extrathyroidal extension showed higher BRAF mutation rate (73.5% vs. 62.3%, P = 0.022). BRAF mutation group showed more aggressive pathologic features, which is considered as higher necessity of radioactive iodine ablation (relative risk, 1.617; P = 0.035). CONCLUSION: This study found that BRAF mutation is associated with classic PTC and central lymph node metastasis and higher necessity of radioactive iodine ablation.

Breast Ultrasound Examination with Video Monitor System: A Satisfaction Survey among Patients

PURPOSE: The purpose of this study is to assess the patients satisfaction with a newly established video-monitor system and the associated basic items for performing breast ultrasound exams by conducting a survey among the patients. MATERIALS AND METHODS: 349 patients were invited to take the survey and they had undergone breast ultrasound examination once during the 3 months after the monitor system has been introduced. The questionnaire was composed of 8 questions, 4 of which were about the basic items such as age, gender and the reason of their taking the breast ultrasound exam, their preference for the gender of the examiner and the desired length of time for the examination. The other 4 question were about their satisfaction with the video monitor. The patients were divided into two groups according to the purposes of taking the exams, which were screening or diagnostic purposes. The results were compared between these 2 groups. The satisfaction with the video monitor system was assessed by using a scoring system that ranged from 1 to 5. RESULTS: For the total patients, the screening group was composed of 124 patients and the diagnostic group was composed of 225. The reasons why the patients wanted to take the examinations in the diagnostic group varied. The questionnaire about the preference of the gender of the examiner showed that 81.5% in the screening group and 79.1% in the diagnostic group preferred a woman doctor. The required, suitable time for the breast ultrasound examination was 5 to 10 minutes or 10 to 15 minutes for about 70% of the patients. The mean satisfaction score for the video monitor system was as high as 3.95 point. The portion of patients in each group who answered over 3 points for their satisfaction with the monitor system was 88.7% and 94.2%, respectively. CONCLUSION: Our study showed that patients preferred 5-15 minutes for the length of the examination time and a female examiner. We also confirmed high patient satisfaction with the video monitor system.

Pulmonary Toxicity after a Quick Course of Combinatorial Vincristine, Bleomycin, and Cisplatin Neoadjuvant Chemotherapy in Cervical Cancer

Pulmonary toxicity is one of the most serious adverse effects associated with a quick course of vincristine, bleomycin, and cisplatin neoadjuvant chemotherapy (NAC-VBP). The aim of this study was to evaluate pulmonary toxicity related to a quick course NAC-VBP. A total of consecutive 61 patients, who underwent at most 3 cycles of NAC-VBP every 10 days in the International Federation of Gynecology and Obstetrics (FIGO) stage IB-IIB cervical cancer from 1995 to 2007, were retrospectively analyzed. Of the 61 study subjects, 7 (11.5%) were identified to have pulmonary toxicity and 2 (3.3%) died of pulmonary fibrosis progression despite aggressive treatment and the use of a multidisciplinary approach. No factor predisposing pulmonary toxicity was identified. Initial symptoms were non-specific, but bronchiolitis obliterans organizing pneumonia and interstitial pneumonitis were characteristic findings by high-resolution computed tomography of the chest. The benefit of steroid therapy was uncertain and was associated with steroid-induced diabetes mellitus requiring insulin therapy in two patients. Fatal pulmonary toxicity is a major concern of a quick course NAC-VBP. In conclusion, these patients require special monitoring for bleomycin-induced pulmonary toxicity.

Fracture Detection with Ultrasonography in Multiple Bony Structures

The diagnosis of a bony fracture by the use of ultrasound is not common. However, with the advantage of the use of a real time examination and recent technical advances, the use of ultrasound has gradually been on the increase. In this pictorial review, we demonstrate the sonographic findings of various kinds of bony fractures where an ultrasound examination was effective in the diagnosis of a fracture. The use of ultrasound is compared with the use of other imaging modalities.

Outcome of Fetal Omphalocele According to Omphalocele Content Combined with Associated Anomaly

PURPOSE: To evaluate the risk of chromosomal abnormality and outcome of fetal omphalocele according to the omphalocele contents combined with associated anomalies. MATERIALS AND METHODS: We retrospectively reviewed the sonograms and case records of 39 fetuses with an omphalocele prenatally detected by ultrasound between 1996 and 2004. We categorized them into two groups according to the omphalocele contents, and the two groups were further subdivided according to the presence or absence of associated anomalies on the prenatal ultrasound. We compared the risk of chromosomal abnormality and fetal outcome according to the omphalocele contents combined with associated anomalies. RESULTS: We categorized the 39 fetuses with an omphalocele into 20 cases with a liver- and bowelcontaining (group I) omphalocele and 19 cases with a bowel-only (group II) omphalocele. Seven of 20 (35%) fetuses with group I omphalocele and 17 of 19 (89%) with group II omphalocele had associated anomalies. Fetal karyotyping was performed in 31 of the 39 fetuses: in group II omphalocele, 13 of the 14 tested fetuses (93%) had chromosomal abnormalities, while in group I omphalocele, one of the 17 tested fetuses (6%) had chromosomal abnormalities. All 13 group II omphalocele with associated anomaly showed abnormal karyotype. Fourteen of 39 (36%) fetuses survived, including 13 fetuses with group I omphalocele (13/20, 65%) and one with group II omphalocele (1/19, 5%). Eleven of 13 (85%) fetuses with isolated group I omphalocele showed good outcome, while no group II omphalocele with associated anomaly survived. CONCLUSION: The bowel-only omphalocele with associated anomalies suggests a very high risk of chromosomal abnormality and a poor outcome. The isolated liver- and bowel-containing omphalocele has a good outcome with a low risk of chromosomal abnormality. Sonographic evaluation of the omphalocele contents and associated anomalies is essential to predict the prognosis of the fetal omphalocele.

Prenatal Sonographic Diagnosis of Focal Musculoskeletal Anomalies

Focal musculoskeletal anomalies vary, and can manifest as part of a syndrome or be accompanied by numerous other conditions such as genetic disorders, karyotype abnormalities, central nervous system anomalies and other skeletal anomalies. Isolated focal musculoskeletal anomaly does, however, also occur; its early prenatal diagnosis is important in deciding prenatal care, and also helps in counseling parents about the postnatal effects of numerous possible associated anomalies. We have encountered 50 cases involving focal musculoskeletal anomalies, including focal limb dysplasia [radial ray abnormality (n=3), mesomelic dysplasia (n=1) ]; anomalies of the hand [polydactyly (n=8), syndactyly (n=3), ectrodactyly (n=1), clinodactyly (n=6), clenched hand (n=5) ]; anomalies of the foot [clubfoot (n=10), rockerbottom foot (n=5), sandal gap deformity (n=1), curly toe (n=2) ]; amniotic band syndrome (n=3) ; and anomalies of the focal spine [block vertebra (n=1), hemivertebra (n=1) ]. Among these 50 cases, five [polydactyly (n=1), syndactyly (n=2) and curly toe (n=2) ] were confirmed by postnatal physical evaluation, two (focal spine anomalies) were diagnosed after postnatal radiologic examination, and the remaining 43 were proven at autopsy. For each condition, we describe the prenatal sonographic findings, and include a brief review.

Congenital Anomalies of the Spine: Radiologic Findings

Congenital anomalies of the spine are frequent and variable. Some are restricted to skeletal structures, while others involve combined neural tube defects or are associated with other multi-systemic disorders. Structural spinal anomalies can be classified according to their location: 1) the vertebral body, 2) the articular process, 3) the lamina with spinous process, 4) the pars interarticularis, 5) the facet joint, 6) the pedicle, or 7) other. Because of similarities between these congenital anomalies and (a) secondary changes involving infection or joint disease and (b) deformities resulting from trauma and uncertain tumorous conditions, significant confusion can occur during diagnosis. Moreover, since the anomalies often give rise to both functional impairment and cosmetic problems, appropriate treatment relies crucially on accurate diagnosis. The authors illustrate the pathogenesis and radiologic findings of the relatively common spinal anomalies confined to skeletal structures.

Anterior Labral Tear: Diagnostic Value of MR Arthrography of the Shoulder

PURPOSE: To assess the accuracy of magnetic resonance(MR) arthrography in the diagnosis of anterior labral tear of the shoulder. MATERIALS AND METHODS: Between Semptember 1996 and February 2000, MR arthrography of the shoulder was performed in 281 patients with a history of shoulder pain or instability. Among this total, only 157 shoulders in 154 patients who underwent arthroscopy or open surgery 0 to 230 (average, 20.9) days after MR arthrography were included in this study; the subjects comprised of 150 males and 4 females with an average age of 23.3 years. MR arthrographs of these 154 patients were analyzed for the presence of anterior labral tears, and the findings were correlated with the arthroscopic and surgical findings. Anterior labral tear was classified as A to D according to its location, as determined by arthroscopy and surgery. (A=4 to 6 o 'clock direction, anteroinferior; B=2 to 4 o 'clock direction, central; C=12 to 2 o 'clock direction, anterosuperior; D= SLAP lesions). The retrospective analysis of MR arthrographs showing false-positive and negative findings was also undertaken. RESULTS: In the diagnosis of anterior labral tear, MR arthrography showed a sensitivity of 94%, a specificity of 90% and an accuracy of 91%. Anterior labral tears were confirmed by arthroscopy or surgery in 62 of the 157 shoulders (39%). Among 62 lesions, two (3%) were observed in area A, 32(52%) in area A+B, nine (15%) in area A+B+C, one(2%) in area A+B+D, 13(21%) in area A+B+C+D, two (3%) in area B+C, one(2%) in area B+D, and two(3%) in area C. Among ten false-positive cases, seven were focal lesions (two, three and two lesions in area A, B and C, respectively), and in the remaining three cases, located in area A+B, MR arthrography revealed thickening and deformation. All four false negatives were focal lesions (two in area A and two in area C). CONCLUSION: Other than in focal lesions, in which accuracy was relatively low, MR arthrography showed high sensitivity, specificity and accuracy in the diagnosis of anterior labral tear of the shoulder.