A deep ensemble medical image segmentation with novel sampling method and loss function

Medical image segmentation is a critical task in computer vision because of facilitating precise identification of regions of interest in medical images. This task plays an important role in disease diagnosis and treatment planning. In recent years, deep learning algorithms have exhibited remarkable performance in this domain. However, it is important to note that there are still unresolved issues, including challenges related to class imbalance and achieving higher levels of accuracy. Considering the challenges, we propose a novel approach to the semantic segmentation of medical images. In this study, a new sampling method to handle class imbalance in the medical datasets is proposed that ensures a comprehensive understanding of both abnormal tissues and background characteristics. Additionally, we propose a novel loss function inspired by exponential loss, which operates at the pixel level. To enhance segmentation performance further, we present an ensemble model comprising two UNet models with ResNet backbone. The initial model is trained on the primary dataset, while the second model is trained on the dataset obtained through our sampling method. The predictions of both models are combined using an ensemble model. We have assessed the effectiveness of our approach using three publicly available datasets: Kvasir-SEG, FLAIR MRI Low-Grade Glioma (LGG), and ISIC 2018 datasets. In our evaluation, we have compared the performance of our loss function against four different loss functions. Furthermore, we have showcased the excellence of our approach by comparing it with various state-of-the-art methods.

Topical Nigella sativa L. product: a new candidate for the management of diabetic peripheral neuropathy

BACKGROUND

Diabetic neuropathy is one of the most common complications of diabetes. The synthetic drugs available in the market have side effects and limitations for diabetic patients, the vast majority of whom are in the upper age group. In this regard, based on Persian medicinal sources, Nigella sativa (N. sativa) has proved to have beneficial effects on neuropathic pain and neurological disorders. In this study, the effect of N. sativa is investigated topically in patients with diabetic neuropathy.

METHODS

This study was performed as a double-blind clinical trial on 120 neuropathic patients. The patients were divided into three groups. The first group received a topical N. sativa product as an ointment, the second group was given a topical placebo, and the third received 300 mg gabapentin capsules. The blindness was done in first and second groups. Diabetic neuropathy was assessed before the study using the Michigan Neuropathy Screening Instrument (MNSI). In addition, neuropathy symptoms were evaluated after the trial using the MNSI questionnaire.

RESULTS

The data were elicited from the patients' answers to a number of questions in the Michigan questionnaire. There were statistically significant differences between the group that received the topical N. sativa product and the other two groups in terms of legs and feet numbness (p value = 0.001), burning pain in feet or legs (p value = 0.001), muscle cramps in feet or legs (p value = 0.001), prickling fleeing in feet or legs (p value = 0.001), hurting of the skin when the bed covers touch it (p value = 0.005), aggravated symptoms at night (p value = 0.001) and hurting feelings in the legs when walking (p value = 0.032). However, the three studied groups were not statistically different in distinguishing hot water from cold water.

CONCLUSION

According to the results of this study, the topical use of N. sativa, compared to the current drugs, has acceptable improving effects on diabetic neuropathic patients.

Acetazolamide and bevacizumab combination therapy versus bevacizumab monotherapy in macular edema secondary to retinal vein occlusion

PURPOSE

To determine and compare the efficacy of intravitreal bevacizumab (IVB) and oral acetazolamide (OA) combination therapy versus IVB monotherapy in patients with macular edema secondary to retinal vein occlusion (RVO).

METHODS

This randomized clinical trial included 54 eyes of 52 patients with RVO central macular thickness (CMT) of more than 300μm, and best corrected visual acuity (BCVA) between 20/400 and 20/40. Eligible patients were randomly assigned to two groups: (I) IVB and OA (250mg twice daily) combination therapy or (II) IVB monotherapy. Ocular injections were repeated monthly for up to three months; BCVA and CMT were measured monthly.

RESULTS

Both regimens resulted in significant reduction in CMT (534±150μm to 352±90μm in the IVB+OA group, P<0.001; and 580±175μm to 362±90μm in the IVB group, P<0.001); neither showed superiority in this regard. Likewise, BCVA showed significant improvement in both groups (0.87±0.56 to 0.53±0.28 LogMAR in the IVB+OA group, P=0.001; and 0.85±0.62 to 0.46±0.4 LogMAR in the IVB group, P<0.001), with no intergroup difference.

CONCLUSION

Addition of oral acetazolamide to IVB in eyes with macular edema secondary to RVO may not result in additional short-term benefits regarding functional and anatomical outcomes.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT05290948, registered on March 22, 2022. https://clinicaltrials.gov/ct2/show/NCT05290948.

Effect of myricetin on the gene expressions of NOX3, TGF-β1, prestin, and HSP-70 and anti-oxidant activity in the cochlea of noise-exposed rats

Objective(s):

Noise-induced hearing loss is one of the most common occupational diseases in industrialized countries and can be affected by various environmental and genetic factors. This study was designed to examine the effect of myricetin in preventing this disorder.

Materials and Methods:

Twenty-one Wistar rats were randomly divided into five groups: Non-exposed, noise exposure only, noise exposure with vehicle, noise exposure with myricetin 5 mg/Kg, and noise exposure with myricetin 10 mg/kg. All animals were sacrificed after last noise exposure. The left cochlea was dissected from each rat. It was used for mRNA expression analysis (NOX3, TGF-β1, prestin, and HSP-70). Blood samples were collected to assess superoxide dismutase (SOD) activity, 1, 1 diphenyl picrylhydrazyl (DPPH), and malondialdehyde (MDA) measurements.

Results:

Real time-PCR assay revealed that noise decreased NOX3 and increased TGF-β1, prestin, and HSP-70 gene expressions. Administration of myricetin at the dose of 5 mg/kg, but not at 10 mg/kg, significantly reversed these changes. Noise also increased MDA levels and decreased SOD and DPPH scavenging activities. Myricetin at the doses of 5 and 10 mg/kg also reversed these changes.

Conclusion:

The findings of this study showed that myricetin at the dose of 5 mg/Kg was able to reverse noise-induced abnormalities in gene expression and oxidant/anti-oxidant balance. It is a possibility that myricetin via enhancement of anti-oxidant activity induced these effects.