Recent advances in deep learning have brought many breakthroughs in medical image analysis by providing more robust and consistent tools for the detection, classification and quantification of patterns in medical images. Specifically, analysis of advanced modalities such as computed tomography (CT) and magnetic resonance imaging (MRI) has benefited most from the data-driven nature of deep learning. This is because the need of knowledge and experience-oriented feature engineering process can be circumvented by automatically deriving representative features from the complex high dimensional medical images with respect to the target tasks. In this paper, we will review recent applications of deep learning in the analysis of CT and MR images in a range of tasks and target organs. While most applications are focused on the enhancement of the productivity and accuracy of current diagnostic analysis, we will also introduce some promising applications which will significantly change the current workflow of medical imaging. We will conclude by discussing opportunities and challenges of applying deep learning to advanced imaging and suggest future directions in this domain.
Classification ; Diagnostic Imaging ; Efficiency ; Learning* ; Magnetic Resonance Imaging*

PURPOSE: To evaluate the impact of temporary internal ureteral stents on the surgical outcomes of dismembered pyeloplasty in children. MATERIALS AND METHODS: The medical records of 70 children (76 renal units) who underwent dismembered pyeloplasty for ureteropelvic junction (UPJ) obstruction at at Asan Medical Center between January 2005 and December 2010 were retrospectively reviewed. We classified the renal units into the stented group (22 renal units) and the nonstented group (54 renal units). Fifty-four of 70 patients were male and their mean age was 2.2+/-3.8 years old. The mean follow-up period was 29.6+/-16.8 months. RESULTS: Sixty-four children had unilateral UPJ obstruction. The mean stent duration was 31.9 days. As shown by evaluation of radiologic images, there were no significant differences between the stented group and the nonstented group during the follow-up period (p>0.05). The mean preoperative and postoperative anteroposterior pelvic diameters (APPDs) of the nonstented group were 31.3 mm and 15.1 mm, respectively (p<0.001). The preoperative and postoperative grades of hydronephrosis were 3.9 and 2.9, respectively (p=0.037). The mean preoperative and postoperative APPDs of the stented group were 36.4 mm and 15.6 mm, respectively (p<0.001). The preoperative and postoperative grades of hydronephrosis were 4 and 3.1, respectively (p<0.001). Repeat obstruction was shown in 4 subjects as a postoperative complication (5.7%). Two children from each group had recurrent UPJ obstruction, with percentages of 3.7% and 9%, respectively (p=0.575). CONCLUSIONS: In a comparison of nonstented and stented groups during pediatric dismembered pyeloplasty for UPJ obstruction, no significant differences were found in the resolution of hydronephrosis or overall postoperative complications.
Child ; Follow-Up Studies ; Humans ; Hydronephrosis ; Male ; Medical Records ; Postoperative Complications ; Retrospective Studies ; Stents ; Ureter ; Ureteral Obstruction

Macrophage infiltration has been observed in the renal biopsy specimens of diabetic nephropathy (DN), and hyperglycemic state stimulates the renal expression of RANTES (regulated upon activation, normal T-cell expressed and secreted) and MCP-1 (monocyte chemoattractant protein-1). Upregulation of RANTES and MCP-1 with infiltrating macrophages may play a crucial role in the development and progression of DN. Genetic polymorphisms of RANTES and its receptors were reported to be independent risk factors for DN. We genotyped single nucleotide polymorphism (SNPs) in the MCP-1 G-2518A, CCR2 G46295A, RANTES C-28G and G-403A in 177 diabetic end-stage renal disease (ESRD) patients and 184 patients without renal involvement (controls) in order to investigate the effects of these SNPs on DN in Korean patients with type 2 DM. There were no differences in the frequencies of SNPs and the distribution of haplotypes of RANTES promoter SNPs between two groups. In conclusion, there were no associations of MCP-1, CCR2 and RANTES promoter SNPs with diabetic ESRD in Korean population. Prospective studies with clearly-defined, homogenous cohorts are needed to confirm the effect of these genetic polymorphisms on DN.
Aged ; Asian Continental Ancestry Group/*genetics ; Chemokine CCL2/*genetics ; Chemokine CCL5/*genetics ; Chi-Square Distribution ; Diabetes Mellitus, Type 2/complications ; Female ; Gene Frequency ; Genotype ; Haplotypes ; Humans ; Kidney Failure, Chronic/ethnology/etiology/*genetics ; Korea ; Male ; Middle Aged ; *Polymorphism, Single Nucleotide ; Promoter Regions, Genetic

BACKGROUND AND OBJECTIVES: Phosphoinositide specific phospholipase C (PLC) plays a pivotal role in the transmembrane signal transduction pathways by catalyzing the hydrolysis of phosphoinositide 4,5-bisphosphate (PIP2) to yield the intracellular second messengers, diacylglycerol (DG) and inositol 1,4,5-trisphosphate (IP3), in response to the interaction of various ligands with the cell surface receptors. The question arises as to the physiological roles of the phosphoinositide second messenger system in the inner ear. The purpose of this study was to determine whether PLCbeta isozymes are present at the cochlea and what portion of cochlea each PLCbeta isozymes are distributed in. MAERIALS AND METHODS: Two methods, an immunohistochemical staining and western blot for PLCbeta isozymes were used in the rat cochlea. Frozen section and surface preparation were prepared for immunohistochemical staining. The PLCbeta isozymes or proteolytic digests were separated by SDS-polyacrylamide gels and then electrophoretically transferred to nitrocellulose membranes. Rabbit polyclonal antibodies raised against four PLCbeta isozymes were used. RESULTS: Each PLCbeta isozymes showed differential expressions in the cochlea. PLCbeta1 immunoreactivity was observed in the inner and outer hair cells and the spiral ganglion cells; PLCbeta2 in the stria vascularis and PLCbeta3 mainly in the inner hair cells. PLCbeta4 was not observed in cochlea. In western blots of rat cochlea extracts, the PLCbeta isozymes stained several bands corresponding to the known molecular weight of PLCbeta monomers, which are probably proteolytic digests. CONCLUSION: These results suggest that differentially localized each PLCbeta isozymes in the cochlea may have specific roles in signal transduction pathway of auditory system.
Animals ; Antibodies ; Blotting, Western ; Cochlea* ; Collodion ; Ear, Inner ; Frozen Sections ; Gels ; Hair ; Hydrolysis ; Inositol 1,4,5-Trisphosphate ; Isoenzymes* ; Ligands ; Membranes ; Molecular Weight ; Phospholipase C beta ; Phospholipases* ; Rats* ; Receptors, Cell Surface ; Second Messenger Systems ; Signal Transduction ; Spiral Ganglion ; Stria Vascularis ; Type C Phospholipases