Deep learning for image-based liver analysis - A comprehensive review focusing on malignant lesions.

Deep learning-based methods, in particular, convolutional neural networks and fully convolutional networks are now widely used in the medical image analysis domain. The scope of this review focuses on the analysis using deep learning of focal liver lesions, with a special interest in hepatocellular carcinoma and metastatic cancer; and structures like the parenchyma or the vascular system. Here, we address several neural network architectures used for analyzing the anatomical structures and lesions in the liver from various imaging modalities such as computed tomography, magnetic resonance imaging and ultrasound. Image analysis tasks like segmentation, object detection and classification for the liver, liver vessels and liver lesions are discussed. Based on the qualitative search, 91 papers were filtered out for the survey, including journal publications and conference proceedings. The papers reviewed in this work are grouped into eight categories based on the methodologies used. By comparing the evaluation metrics, hybrid models performed better for both the liver and the lesion segmentation tasks, ensemble classifiers performed better for the vessel segmentation tasks and combined approach performed better for both the lesion classification and detection tasks. The performance was measured based on the Dice score for the segmentation, and accuracy for the classification and detection tasks, which are the most commonly used metrics.

Interventions for High-Burden Infectious Diseases in Children and Adolescents: A Meta-analysis.

BACKGROUND: Approximately 2.2 million deaths were reported among school-age children and young people in 2019, and infectious diseases remain the leading causes of morbidity and mortality, especially in low and middle-income countries. We aim to synthesize evidence on interventions for high-burden infectious diseases among children and adolescents aged 5 to 19 years. METHODS: We conducted a comprehensive literature search until December 31, 2020. Two review authors independently screened studies for relevance, extracted data, and assessed risk of bias. RESULTS: We included a total of 31 studies, including 81 596 participants. Sixteen studies focused on diarrhea; 6 on tuberculosis; 2 on human immunodeficiency virus; 2 on measles; 1 study each on acute respiratory infections, malaria, and urinary tract infections; and 2 studies targeted multiple diseases. We did not find any study on other high burden infectious diseases among this age group. We could not perform meta-analysis for most outcomes because of variances in interventions and outcomes. Findings suggests that for diarrhea, water treatment, water filtration, and zinc supplementation have some protective effect. For tuberculosis, peer counseling, contingency contract, and training of health care workers led to improvements in tuberculosis detection and treatment completion. Continuation of cotrimoxazole therapy reduced the risk of tuberculosis and hospitalizations among human immunodeficiency virus-infected children and reduced measles complications and pneumonia cases among measles-infected children. Zinc supplementation led to a faster recovery in urinary tract infections with a positive effect in reducing symptoms. CONCLUSIONS: There is scarcity of data on the effectiveness of interventions for high-burden infectious diseases among school-aged children and adolescents.

COVID-19-Induced Downsizing and Survivors' Syndrome: The Moderating Role of Transformational Leadership.

Downsizing due to COVID-19 (COV-DS) and its consequences on laid-off employees has attracted the attention of many researchers, around the globe. However, the underlying mechanisms that explain the effects of COVID-19 downsizing (COV-DS) on the employees who have survived cutoffs remain underexplored. Grounded in the conservation of resources theory, this manuscript aims to study the causal path through which COV-DS reduces the survivors' affective commitment. The current study proposes the mediation of survivors' job uncertainty, stress, and organizational identification between COV-DS and survivors' affective commitment. This study also posits the moderating role of transformational leadership between COV-DS and both the mediators. The extant study has employed WARPED partial least square WARP PLS 7 and Hayes Process Macro to test the hypothesized relationships. Using the sample of 274 employees from the private sector of Pakistan, it was found that job uncertainty's stress strongly mediates the relationship between COV-DS and survivors' affective commitment. While mediation of survivors' organizational identification was not proven to be significant. However, with the moderation of transformational leadership, both the mediators were proven to be significant.

The Impact of the COVID-19 Pandemic on Immunization Campaigns and Programs: A Systematic Review.

The COVID-19 pandemic has had an impact on health service delivery, including immunization programs, and this review assesses the impact on vaccine coverage across the globe and identifies the potential underlying factors. A systematic search strategy was employed on PubMed, Embase, MedRxiv, BioRxiv, and WHO COVID-19 databases from December 2019 till 15 September 2020. Two review authors independently assessed studies for inclusion, assessed quality, and extracted the data (PROSPERO registration #CRD42020182363). A total of 17 observational studies were included. The findings suggest that there was a reduction in the vaccination coverage and decline in total number of vaccines administered, which led to children missing out on their vaccine doses. An approximately fourfold increase was also observed in polio cases in polio endemic countries. Factors contributing to low vaccine coverage included fear of being exposed to the virus at health care facilities, restriction on city-wide movements, shortage of workers, and diversion of resources from child health to address the pandemic. As the world re-strategizes for the post-2020 era, we should not let a crisis go to waste as they provide an opportunity to establish guidelines and allocate resources for future instances. High-quality supplementary immunization activities and catch-up programs need to be established to address gaps during the pandemic era.

Fast parallel vessel segmentation.

BACKGROUND AND OBJECTIVE: Accurate and fast vessel segmentation from liver slices remain challenging and important tasks for clinicians. The algorithms from the literature are slow and less accurate. We propose fast parallel gradient based seeded region growing for vessel segmentation. Seeded region growing is tedious when the inter connectivity between the elements is unavoidable. Parallelizing region growing algorithms are essential towards achieving real time performance for the overall process of accurate vessel segmentation. METHODS: The parallel implementation of seeded region growing for vessel segmentation is iterative and hence time consuming process. Seeded region growing is implemented as kernel termination and relaunch on GPU due to its iterative mechanism. The iterative or recursive process in region growing is time consuming due to intermediate memory transfers between CPU and GPU. We propose persistent and grid-stride loop based parallel approach for region growing on GPU. We analyze static region of interest of tiles on GPU for the acceleration of seeded region growing. RESULTS: We aim fast parallel gradient based seeded region growing for vessel segmentation from CT liver slices. The proposed parallel approach is 1.9x faster compared to the state-of-the-art. CONCLUSION: We discuss gradient based seeded region growing and its parallel implementation on GPU. The proposed parallel seeded region growing is fast compared to kernel termination and relaunch and accurate in comparison to Chan-Vese and Snake model for vessel segmentation.

GPU acceleration of liver enhancement for tumor segmentation.

BACKGROUND AND OBJECTIVE: Medical image segmentation plays a vital role in medical image analysis. There are many algorithms developed for medical image segmentation which are based on edge or region characteristics. These are dependent on the quality of the image. The contrast of a CT or MRI image plays an important role in identifying region of interest i.e. lesion(s). In order to enhance the contrast of image, clinicians generally use manual histogram adjustment technique which is based on 1D histogram specification. This is time consuming and results in poor distribution of pixels over the image. Cross modality based contrast enhancement is 2D histogram specification technique. This is robust and provides a more uniform distribution of pixels over CT image by exploiting the inner structure information from MRI image. This helps in increasing the sensitivity and accuracy of lesion segmentation from enhanced CT image. The sequential implementation of cross modality based contrast enhancement is slow. Hence we propose GPU acceleration of cross modality based contrast enhancement for tumor segmentation. METHODS: The aim of this study is fast parallel cross modality based contrast enhancement for CT liver images. This includes pairwise 2D histogram, histogram equalization and histogram matching. The sequential implementation of the cross modality based contrast enhancement is computationally expensive and hence time consuming. We propose persistence and grid-stride loop based fast parallel contrast enhancement for CT liver images. We use enhanced CT liver image for the lesion or tumor segmentation. We implement the fast parallel gradient based dynamic seeded region growing for lesion segmentation. RESULTS: The proposed parallel approach is 104.416 ( ±  5.166) times faster compared to the sequential implementation and increases the sensitivity and specificity of tumor segmentation. CONCLUSION: The cross modality approach is inspired by 2D histogram specification which incorporates spatial information existing in both guidance and input images for remapping the input image intensity values. The cross modality based liver contrast enhancement improves the quality of tumor segmentation.

Weakly-Supervised Network for Detection of COVID-19 in Chest CT Scans.

Deep Learning-based chest Computed Tomography (CT) analysis has been proven to be effective and efficient for COVID-19 diagnosis. Existing deep learning approaches heavily rely on large labeled data sets, which are difficult to acquire in this pandemic situation. Therefore, weakly-supervised approaches are in demand. In this paper, we propose an end-to-end weakly-supervised COVID-19 detection approach, ResNext+, that only requires volume level data labels and can provide slice level prediction. The proposed approach incorporates a lung segmentation mask as well as spatial and channel attention to extract spatial features. Besides, Long Short Term Memory (LSTM) is utilized to acquire the axial dependency of the slices. Moreover, a slice attention module is applied before the final fully connected layer to generate the slice level prediction without additional supervision. An ablation study is conducted to show the efficiency of the attention blocks and the segmentation mask block. Experimental results, obtained from publicly available datasets, show a precision of 81.9% and F1 score of 81.4%. The closest state-of-the-art gives 76.7% precision and 78.8% F1 score. The 5% improvement in precision and 3% in the F1 score demonstrate the effectiveness of the proposed method. It is worth noticing that, applying image enhancement approaches do not improve the performance of the proposed method, sometimes even harm the scores, although the enhanced images have better perceptual quality.

Acute Myeloid Leukemia: Is That All There Is?

Acute myeloid leukemia (AML) is characterized by the clonal proliferation of malignant myeloid blast cells in the marrow along with impaired normal hematopoiesis. With an almost stagnant approach for the management of patients with AML in the last three decades, the main purpose of this paper is to increase our understanding of recent scientific advancements for the enhanced diagnosis and treatment of AML. Existing research data related to different approaches for a possible improvement in AML management has been collected and discussed. The identification of recurrently mutated genes, such as CCAAT-enhancer-binding proteins α (CEBPα), Fms-related tyrosine kinase 3 (FLT3), and nucleophosmin 1 (NPM1) along with the classic diagnostic karyotype has improved prognostic-risk stratification. Moreover, mutations affecting cellular metabolism like isocitrate dehydrogenase (IDH1), lysine-specific demethylase 1 (LSD 1), and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) have become a huge success by providing targets for novel therapeutic drugs. Checkpoint inhibitors (CPI) and vaccination against tumor-associated antigen are added options considered, which require further trials before their efficacy can be determined. An important tool in monitoring early response to therapy, minimal residual disease (MRD) assays can be further refined by including pretreatment parameters such as cytogenetic and molecular markers. Potential side effects and resistance to treatment remains a huge barrier in completely finding success against AML and work needs to be done to find combinations of immunotherapies to possibly reduce adaptive resistance by AML.

Prognostic value of blood count parameters in patients with acute coronary syndrome.

BACKGROUND: Recent studies have shown that complete blood count (CBC) parameters can effectively predict long-term mortality and re-infarction rates in acute coronary syndrome (ACS). However, the role of these parameters in predicting short term mortality has not been studied extensively. The main objective of this study was to determine whether CBC parameters can predict 30-days mortality and the incidence of major adverse cardiac event (MACE) in ACS patients. METHODOLOGY: A total of 297 patients with ACS were recruited in this prospective study. The relationship of baseline white blood cell (WBC) to mean platelet volume ratio (WMR) with MACE and mortality was assessed during a 30-days follow up. The patients were divided into two groups: Group A [WMR<1000] and Group B [WMR>1000]. Multivariate COX regression was performed to calculate hazard ratios (HR). RESULTS: WMR had the highest area under receiver operating characteristics curve and highest discriminative ability amongst all CBC parameters in predicting mortality. Patients in Group B had a higher mortality rate (p<0.001) than patients in Group A. WBC count (p=0.02), platelet count (p=0.04), WMR (p=0.008), platelet to lymphocyte ratio (p<0.001) and neutrophil to lymphocyte ratio (p=0.03) were significantly higher in the MACE-positive group as compared to MACE-negative. In multivariate cox regression analysis, WMR>1000 (HR=2.9, 95% confidence interval 1.3-6.5, p=0.01) was found to be strongest biochemical marker in predicting mortality. CONCLUSION: WMR is an easily accessible and an inexpensive indicator, which may be used as a prognostic marker in patients with ACS.

Prevalence and characteristics of resistant hypertensive patients in an Asian population.

BACKGROUND: Resistant hypertension is a well-recognized clinical challenge yet there are no reported data on its prevalence in Pakistan. These patients are subjected to a higher risk of developing hypertensive complications. The objective of our study was to evaluate the prevalence and determinants of resistant hypertension in an Asian cohort of hypertensive patients. METHODS: This cross-sectional study was carried out among hypertensive patients visiting a tertiary care hospital in Karachi from September-December 2015. Patient data and characteristics were recorded using a pre-coded questionnaire. Morisky and Berlin questionnaires were used to assess compliance to medications and determine the risk of developing obstructive sleep apnea, respectively. Pearson's chi-square test was used to analyze statistical differences between hypertensive patients and related factors. RESULTS: A total of 515 patients were included in the study. Overall, 12% of the total patients (n=62) were resistant hypertensives and 25% (n=129) had pseudo-resistant hypertension. Resistant patients were more often females, older and had a higher body mass index (all P<0.001). Use of painkillers and noncompliance to dietary recommendations were found to be significant determinants of resistant hypertension. Prevalence of comorbid conditions, including diabetes (p=0.33), hyperlipidemia (p=0.46), and chronic kidney disease (p=0.23), was not significantly higher in patients with resistant hypertension. CONCLUSION: Nearly one in ten hypertensive patients had true resistant hypertension, and twenty-five percent of patients had pseudo-resistance. Resistance hypertensions is significantly associated with female gender, older age, obesity, dietary noncompliance and increased use of NSAIDs.