ABSTRACT
RATIONALE AND OBJECTIVES: Artificial intelligence (AI) systems have been increasingly applied to breast ultrasonography. They are expected to decrease the workload of radiologists and to improve diagnostic accuracy. The aim of this study is to evaluate the performance of an AI system for the BI-RADS category assessment in breast masses detected on breast ultrasound. MATERIALS AND METHODS: A total of 715 masses detected in 530 patients were analyzed. Three breast imaging centers of the same institution and nine breast radiologists participated in this study. Ultrasound was performed by one radiologist who obtained two orthogonal views of each detected lesion. These images were retrospectively reviewed by a second radiologist blinded to the patient's clinical data. A commercial AI system evaluated images. The level of agreement between the AI system and the two radiologists and their diagnostic performance were calculated according to dichotomic BI-RADS category assessment. RESULTS: This study included 715 breast masses. Of these, 134 (18.75%) were malignant, and 581 (81.25%) were benign. In discriminating benign and probably benign from suspicious lesions, the agreement between AI and the first and second radiologists was moderate statistically. The sensitivity and specificity of radiologist 1, radiologist 2, and AI were calculated as 98.51% and 80.72%, 97.76% and 75.56%, and 98.51% and 65.40%, respectively. For radiologist 1, the positive predictive value (PPV) was 54.10%, the negative predictive value (NPV) was 99.58%, and the accuracy was 84.06%. Radiologist 2 achieved a PPV of 47.99%, NPV of 99.32%, and accuracy of 79.72%. The AI system exhibited a PPV of 39.64%, NPV of 99.48%, and accuracy of 71.61%. Notably, none of the lesions categorized as BI-RADS 2 by AI were malignant, while 2 of the lesions classified as BI-RADS 3 by AI were subsequently confirmed as malignant. By considering AI-assigned BI-RADS 2 as safe, we could potentially avoid 11% (18 out of 163) of benign lesion biopsies and 46.2% (110 out of 238) of follow-ups. CONCLUSION: AI proves effective in predicting malignancy. Integrating it into the clinical workflow has the potential to reduce unnecessary biopsies and short-term follow-ups, which, in turn, can contribute to sustainability in healthcare practices.
ABSTRACT
BACKGROUND: Detachment and embolization (DE) is a rare complication of totally implantable central venous access devices (TIVADs). This study aimed to analyze clinical findings, etiology, and treatment options in DE of TIVADs. METHODS: Patients who experienced DE between 2010-2019 were included. Indications, implantation techniques, time to diagnosis, patient complaints, diagnostic methods, rupture site, location of embolization, treatment methods, and chest X-rays prior to detachment were analyzed retrospectively. RESULTS: DE of TIVAD was detected in 12(1.2%) patients. Eleven patients had breast cancer and one had colon cancer. Mean age at implantation was 45.3 ± 9.6(31-61.3) years. Seven (58%) patients were asymptomatic, four (33.3%) had TIVAD malfunction, and one (8.3%) had pain and swelling at port site after injection. Mean time from implantation to diagnosis was 1149.92(16-2795) days. The etiologies comprised Pinch-off Syndrome (POS) in eight (66%) patients, detachment directly adjacent to the lock mechanism in three (25%) patients, and probable iatrogenic injury during explantation in one (9%) patient. The most common site of embolism was the superior vena cava (25%). While the embolized fragment was removed percutaneously in 11 patients, medical follow-up was treatment choice for one patient. CONCLUSIONS: DE is a rare complication with an incidence rate of 1.2% in this study. Since most patients were asymptomatic, chest radiography plays an important role in diagnosis. The most common cause was POS, and it can be prevented by inserting the catheter from lateral third of the clavicle during subclavian vein catheterization. The first-choice treatment was percutaneous femoral retrieval. However, if not technically possible, alternative treatment options are thoracotomy or follow-up with anticoagulant therapy.
Subject(s)
Catheterization, Central Venous , Catheterization, Central Venous/adverse effects , Catheters, Indwelling/adverse effects , Device Removal , Humans , Retrospective Studies , Syndrome , Vena Cava, SuperiorABSTRACT
Localized amyloidosis in the breast is a very rare disease and may mimic malignant lesions. A 60-year-old woman who had a history of breast-conserving surgery presents with a new a well-defined oval opacity accompanied by many round tight clustered micro- and macrocalcifications on mammograms. It could not be visualized sonographically due to the intense posterior acoustic shadowing of the fat necrosis areas and contrast enhancement was not detected in this area on the dynamic contrast enhanced magnetic resonance images. At pathological examination breast amyloidosis was detected. Amyloidosis of the breast is a rare disease, but it can mimic malignancy and should be included in the differential diagnosis.
ABSTRACT
PURPOSE: To determine the contribution of apparent diffusion coefficient (ADC), and relative ADC (rADC) values to differentiate between benign and malignant breast masses. MATERIALS AND METHODS: Magnetic resonance imaging (MRI) of the breast with diffusion-weighted imaging (DWI) of patients with benign or malignant breast masses diagnosed either by histopathological findings or by follow-up imaging were evaluated retrospectively. Histopathological analyses were performed for 71 lesions (80.7%) while the remaining were followed up every six months for one year. DWI was performed using b-values of 0 and 1000 sec/mm2, and ADC and rADC were calculated and compared. A receiver operating characteristic (ROC) curve and Youden index were used to evaluate the parameter's optimal threshold and diagnostic value. Statistical significance was set as p < 0.05. RESULTS: Eighty-eight lesions from a total of 81 patients, aged between 16 and 73 (mean age 42 ± 11.3) years were obtained and evaluated. Pathological results of 34 (38.6%) out of 71 lesions were malignant and 37 lesions (42%) were benign. Seventeen (19.3%) lesions remained stable at one-year follow-up and were accepted as benign breast masses. Mean ADC values of benign and malignant lesions were 1.584 × 10-3mm2/sec and 0.884 × 10-3mm2/sec (p < 0.05), respectively. Sensitivity and specificity of ADC were 88% and 87%, respectively at a cut-off value of 1.04 × 10-3mm2/sec. Mean rADC was 0.931 for benign lesions and 0.557 for malignant lesions (p < 0.05). Sensitivity and specificity were 82% and 83% at a cut-off value of 0.639. No prominent superiority of rADC over ADC is identified in the differentiation of breast masses. CONCLUSION: ADC and rADC values derived from DWI can be equally useful in clinical setting to differentiate benign from malignant breast masses.
