MyMedR

Displaying all 4 publications

Abstract:

Sort:

Absorbable versus nonabsorbable sutures for subcuticular skin closure of a transverse suprapubic incision

Tan PC, Mubarak S, Omar SZ

Int J Gynaecol Obstet, 2008 Nov;103(2):179-81.
PMID: 18639876 DOI: 10.1016/j.ijgo.2008.05.023
Gamma-glutamyltransferase level in pregnancy is an independent risk factor for gestational diabetes mellitus

Tan PC, Mubarak S, Omar SZ

J Obstet Gynaecol Res, 2008 Aug;34(4):512-7.
PMID: 18937705 DOI: 10.1111/j.1447-0756.2008.00815.x

AIM:
To evaluate the relationship between gamma-glutamyltransferase (GGT) level in pregnant women at oral glucose tolerance test (OGTT) and the diagnosis of gestational diabetes (GDM).

METHODS:
Blood was taken for analyzing GGT level from women at high risk of GDM at the time of their scheduled OGTT. GDM was diagnosed according to World Health Organization 1999 criteria.

RESULTS:
GGT level correlated positively with the 2-hour glucose level (Spearman's rho = 0.112: P < 0.05). GGT values that were stratified into quartiles demonstrated a significant trend with diagnosis of GDM (chi(2) for trend; P = 0.03). Multivariable logistic regression analysis taking into account maternal age, gestational age at OGTT, body mass index and a positive 50-g glucose challenge test (GCT) indicated that high GGT was an independent risk factor for GDM (adjusted odds ratio [AOR] 2.1 95% CI 1.2-3.8: P = 0.01). In the subset of women identified by a positive GCT, on multivariable logistic regression analysis, only high GGT was an independent risk factor for GDM (AOR 2.3 95% CI 1.3-4.2: P = 0.007).

CONCLUSION:
Raised GGT level is an independent risk factor for GDM in high risk pregnant women undergoing OGTT.
Fulltext Ultrasound guidance versus the blind method for intrauterine catheter insemination: A randomized controlled trial

Mubarak S, Yusoff NH, Adnan TH

Clin Exp Reprod Med, 2019 Jun;46(2):87-94.
PMID: 31181876 DOI: 10.5653/cerm.2019.46.2.87

OBJECTIVE: The primary objective of this study was to compare clinical pregnancy rates in intrauterine insemination (IUI) treatment cycles with transabdominal ultrasound guidance during intrauterine catheter insemination (US-IUI) versus the "blind method" IUI without ultrasound guidance (BM-IUI). The secondary objective was to compare whether US-IUI had better patient tolerability and whether US-IUI made the insemination procedure easier for the clinician to perform compared to BM-IUI.
METHODS: This was a randomized controlled trial done at the Reproductive Medicine Unit of General Hospital Kuala Lumpur, Malaysia. We included women aged between 25 and 40 years who underwent an IUI treatment cycle with follicle-stimulating hormone injections for controlled ovarian stimulation.
RESULTS: A total of 130 patients were recruited for our study. The US-IUI group had 70 patients and the BM-IUI group had 60 patients. The clinical pregnancy rate was 10% in both groups (p> 0.995) and there were no significant difference between the groups for patient tolerability assessed by scores on a pain visual analog scale (p= 0.175) or level of difficulty for the clinician (p> 0.995). The multivariate analysis further showed no significant increase in the clinical pregnancy rate (adjusted odds ratio, 1.07; 95% confidence interval, 0.85-1.34; p= 0.558) in the US-IUI group compared to the BM-IUI group even after adjusting for potential covariates.
CONCLUSION: The conventional blind method for intrauterine catheter insemination is recommended for patients undergoing IUI treatment. The use of ultrasound during the insemination procedure increased the need for trained personnel to perform ultrasonography and increased the cost, but added no extra benefits for patients or clinicians.
Fulltext A generative adversarial network for synthetization of regions of interest based on digital mammograms

Oyelade ON, Ezugwu AE, Almutairi MS, Saha AK, Abualigah L, Chiroma H

Sci Rep, 2022 Apr 13;12(1):6166.
PMID: 35418566 DOI: 10.1038/s41598-022-09929-9

Deep learning (DL) models are becoming pervasive and applicable to computer vision, image processing, and synthesis problems. The performance of these models is often improved through architectural configuration, tweaks, the use of enormous training data, and skillful selection of hyperparameters. The application of deep learning models to medical image processing has yielded interesting performance, capable of correctly detecting abnormalities in medical digital images, making them surpass human physicians. However, advancing research in this domain largely relies on the availability of training datasets. These datasets are sometimes not publicly accessible, insufficient for training, and may also be characterized by a class imbalance among samples. As a result, inadequate training samples and difficulty in accessing new datasets for training deep learning models limit performance and research into new domains. Hence, generative adversarial networks (GANs) have been proposed to mediate this gap by synthesizing data similar to real sample images. However, we observed that benchmark datasets with regions of interest (ROIs) for characterizing abnormalities in breast cancer using digital mammography do not contain sufficient data with a fair distribution of all cases of abnormalities. For instance, the architectural distortion and breast asymmetry in digital mammograms are sparsely distributed across most publicly available datasets. This paper proposes a GAN model, named ROImammoGAN, which synthesizes ROI-based digital mammograms. Our approach involves the design of a GAN model consisting of both a generator and a discriminator to learn a hierarchy of representations for abnormalities in digital mammograms. Attention is given to architectural distortion, asymmetry, mass, and microcalcification abnormalities so that training distinctively learns the features of each abnormality and generates sufficient images for each category. The proposed GAN model was applied to MIAS datasets, and the performance evaluation yielded a competitive accuracy for the synthesized samples. In addition, the quality of the images generated was also evaluated using PSNR, SSIM, FSIM, BRISQUE, PQUE, NIQUE, FID, and geometry scores. The results showed that ROImammoGAN performed competitively with state-of-the-art GANs. The outcome of this study is a model for augmenting CNN models with ROI-centric image samples for the characterization of abnormalities in breast images.