This paper performs a comprehensive study on the deep-learning-based computer-aided diagnosis (CADx) for the differential diagnosis of benign and malignant nodules/lesions by avoiding the potential errors caused by inaccurate image processing results (e.g., boundary segmentation), as well as the classification bias resulting from a less robust feature set, as involved in most conventional CADx algorithms. Specifically, the stacked denoising auto-encoder (SDAE) is exploited on the two CADx applications for the differentiation of breast ultrasound lesions and lung CT nodules. The SDAE architecture is well equipped with the automatic feature exploration mechanism and noise tolerance advantage, and hence may be suitable to deal with the intrinsically noisy property of medical image data from various imaging modalities. To show the outperformance of SDAE-based CADx over the conventional scheme, two latest conventional CADx algorithms are implemented for comparison. 10 times of 10-fold cross-validations are conducted to illustrate the efficacy of the SDAE-based CADx algorithm. The experimental results show the significant performance boost by the SDAE-based CADx algorithm over the two conventional methods, suggesting that deep learning techniques can potentially change the design paradigm of the CADx systems without the need of explicit design and selection of problem-oriented features.

The overall percentage of missed fractures in the extremities was 3.7%. Only 33% of the initially missed fractures were attributed to radiographically imperceptible lesions. Adequate training for physicians and radiologists in the ED may reduce the rate of missed fractures.

No abstract

Acute urinary retention is unusual in children and is usually a candidate for visiting the emergency department upon initial discovery. We report a 12-year-old girl who complained of acute urinary retention. Ultrasonography demonstrated a large echogenic mass over the vagina and mild dilation of the uterus. Imperforate hymen associated with hematocolpos and hematometrium was diagnosed. Cruciate hymenotomy was performed. The symptoms resolved after treatment. Adolescent girls who complain of urinary symptoms with no previous menstruation should have their external genitalia examined in order to rule out the possibility of imperforate hymen as the cause of acute urinary retention.

During a recent 5-year period, 12 patients with splenic abscesses were evaluated by abdominal ultrasound (US) examination. Multifocal abscesses were noted in seven patients, three of them were secondary to infectious endocarditis, three were in immunosuppressed state, and one was caused by tuberculosis. The latter four patients had developed splenic microabscesses with a diameter of less than 1.5 cm. The larger abscesses showed an irregular wall, weak or no internal echoes, ovoid or round in shape, and accompanied by mild to moderate distal acoustic enhancement. Wedge-shaped abscesses were typically noted in patients with infectious endocarditis and septic embolism. US-guided percutaneous drainage was done in five patients (abscesses greater than 4 cm). Simple aspiration in conjunction with antibiotic administration was done for seven smaller abscesses (diameter less than 3.5 cm) in five patients. A second drainage, either for a dislodged catheter or a recurrent abscess, was performed in two cases. All patients had uneventful clinical course following this therapeutic approach.