Writing mammography reports can be error-prone and time-consuming for radiologists. In this paper we propose a method to generate mammography reports given four images, corresponding to the four views used in screening mammography. To the best of our
knowledge our work represents the first attempt to generate the mammography report using deep-learning. We propose an encoder-decoder model that includes an EfficientNet-based encoder and a Transformer-based decoder. We demonstrate that the Transformer-based attention mechanism can combine visual and semantic information to localize salient regions on the input mammograms and generate a visually interpretable report. The conducted experiments, including an evaluation by a certified radiologist, show the effectiveness of the proposed method.
Background and objectives: The advent of high-resolution MRI with a dedicated epilepsy protocol improves the ability to identify possible structural abnormalities that underlie seizure disorders. The aims of this study were to evaluate the diagnostic
efficacy of standard MRI, identify whether there is an increase in the diagnostic yield with the addition of dedicated seizure protocol, and compare the diagnostic yields of MRI and electroencephalogram (EEG) individually and in combination. Subjects and Methods: This was a cross-sectional analytic study, included 100 cases who presented with seizure over 18 months. Patients underwent complete neurological examination, EEG, and MRI with a standard and dedicated epilepsy protocol. Results: We found epileptogenic lesions in MRI in 55.5. Mesial temporal lobe sclerosis was the most common epileptogenic lesion (45.5%). The diagnostic efficacy of MRI had increased with dedicated epilepsy protocol compared to standard protocol. Abnormal MRI and EEG were compatible in 21%. Conclusion and implications: Dedicated epilepsy protocol increased the diagnostic efficacy of brain MRI in detecting a structural epileptogenic lesion, with 100% of mesial temporal sclerosis, the most common lesion in our study, was detected only in dedicated epilepsy protocol and missed in standard protocol.
112 patients with suspected breast lesions were investigated using Scintimammography with 99mTc-MIBI, mammography and magnetic resonance to compare the diagnostic accuracy of Scintimammography with that of mammography and magnetic resonance in the d
etecting of primary breast cancer. Excisional biopsy was taken after 7-10 days of Scintimammography. 70 patients were diagnosed with breast cancer; 55 palpable and 15 non-palpable lesions, while 42, 30 palpable and 12 non-palpable were diagnosed as benign lesions. The sensitivity of the three methods were high and approximate (89% by Scintimammography, 90% by mammography and 94% by magnetic resonance), while Scintimammography showed higher specificity (86%) versus 21% for mammography and 50% for magnetic resonance. The high specificity in addition to high positive (91%) and negative (82%) predictive values of Scintimammography resulted in a highly accurate diagnostic (87%) method compared to mammography (64%) and magnetic resonance (78%) procedures. Through the complementary use of Scintimammography it is possible to increase the sensitivity for the detection of breast cancer. In patients in whom the status of a palpable breast mass remains unclear, Scintimammography may help to reduce the amount of unnecessary biopsies.
The aim of this study was to identify the configuration of the mandibular canal
radiographically in the vertical direction on Digital Panoramic Radiographs to evaluate the relationship
of the mandibular canal with the Status of the third mandibular Molar which will add important
diagnostic radiographic signs specially before performing extraction of the mandibular 3rd molar .
Background& Objective: The reduction of alveolar crestal bone density is one of the earlier signs of periimplant
disease and precedes the loss of height of the alveolar crest, because of that early detection of the
small changes in density is impor
tant and lead to the early onset of treatment and increases its success's
percentage.
This study aimed to evaluate the ability of the direct conventional digital method in detection of density
changes around dental implants by the mean value of the gray level of the histogram of the selected area
comparing with radiographic subtraction method.
