An Efficient Multi-Scale Fusion Network for 3D Organs at Risk (OARs) Segmentation.

Accurate segmentation of organs-at-risks (OARs) is a precursor for optimizing radiation therapy planning. Existing deep learning-based multi-scale fusion architectures have demonstrated a tremendous capacity for 2D medical image segmentation. The key to their success is aggregating global context and maintaining high resolution representations. However, when translated into 3D segmentation problems, existing multi-scale fusion architectures might underperform due to their heavy computation overhead and substantial data diet. To address this issue, we propose a new OAR segmentation framework, called OARFocalFuseNet, which fuses multi-scale features and employs focal modulation for capturing global-local context across multiple scales. Each resolution stream is enriched with features from different resolution scales, and multi-scale information is aggregated to model diverse contextual ranges. As a result, feature representations are further boosted. The comprehensive comparisons in our experimental setup with OAR segmentation as well as multi-organ segmentation show that our proposed OARFocalFuseNet outperforms the recent state-of-the-art methods on publicly available OpenKBP datasets and Synapse multi-organ segmentation. Both of the proposed methods (3D-MSF and OARFocalFuseNet) showed promising performance in terms of standard evaluation metrics. Our best performing method (OARFocalFuseNet) obtained a dice coefficient of 0.7995 and hausdorff distance of 5.1435 on OpenKBP datasets and dice coefficient of 0.8137 on Synapse multi-organ segmentation dataset. Our code is available at https://github.com/NoviceMAn-prog/OARFocalFuse.

A Population Model of Time-Dependent Changes in Serum Creatinine in (Near)term Neonates with Hypoxic-Ischemic Encephalopathy During and After Therapeutic Hypothermia.

The objective was to apply a population model to describe the time course and variability of serum creatinine (sCr) in (near)term neonates with moderate to severe encephalopathy during and after therapeutic hypothermia (TH). The data consisted of sCr observations up to 10 days of postnatal age in neonates who underwent TH during the first 3 days after birth. Available covariates were birth weight (BWT), gestational age (GA), survival, and acute kidney injury (AKI). A previously published population model of sCr kinetics in neonates served as the base model. This model predicted not only sCr but also the glomerular filtration rate normalized by its value at birth (GFR/GFR0). The model was used to compare the TH neonates with a reference full term non-asphyxiated population of neonates. The estimates of the model parameters had good precision and showed high between subject variability. AKI influenced most of the estimated parameters denoting a strong impact on sCr kinetics and GFR. BWT and GA were not significant covariates. TH transiently increased [Formula: see text] in TH neonates over the first days compared to the reference group. Asphyxia impacted not only GFR, but also the [Formula: see text] synthesis rate. We also observed that AKI neonates exhibit a delayed onset of postnatal GFR increase and have a higher [Formula: see text] synthesis rate compared to no-AKI patients. Our findings show that the use of [Formula: see text] as marker of renal function in asphyxiated neonates treated with TH to guide dose selection for renally cleared drugs is challenging, while we captured the postnatal sCr patterns in this specific population.

Comparison of Two Different Positions for Ultrasound-Guided Intervertebral Distance Evaluation.

Objective: During neuraxial anaesthesia, correct patient positioning is key for increased block success and (patient) comfort. The aim of this prospective study was to compare the lateral fetal decubitus (LFD) position with the sitting fetal lotus (SFL) regarding interspinous distance, transverse diameters of paravertebral muscles measured with ultrasonography, and patient comfort. Methods: Fifty adult participants who could sit cross-legged and had no lumbar anomalies were included in our prospective study. In both SFL and LFD positions, measurements were performed with ultrasonography; in the axial plane, interspinous distance at the level of L4-L5, in the sagittal plan, with the probe slightly tilted, subcutaneous tissue-spinous process depth, and transverse diameters of paravertebral muscles were measured. Stretcher, waist position, and abdominal comfort were scored on a scale of 1 (very bad) to 7 (perfect) with a verbal numeric satisfaction scale. Results: Interspinous distance was significantly larger in the SFL position than in the LFD position (P < 0.05). There was no significant difference between the two positions (P > 0.05) regarding patient comfort. Paravertebral muscle diameters were significantly broader in the SFL position than in the LFD position. The diameter of the left paravertebral muscle in the SFL position (45.8±8.8 mm) was larger than that in the LFD position (43±7.8 mm; P < 0.001). The diameter of the right paravertebral muscle in the SFL position was (47±9 mm) larger than that in the LFD position (43.4±7.6 mm; P < 0.001). Conclusion: Although there was no difference regarding the comfort between the two positions, the interspinous distance was larger in the SFL position than in the LFD position.

The past, current, and future of neonatal intensive care units with artificial intelligence: a systematic review.

Machine learning and deep learning are two subsets of artificial intelligence that involve teaching computers to learn and make decisions from any sort of data. Most recent developments in artificial intelligence are coming from deep learning, which has proven revolutionary in almost all fields, from computer vision to health sciences. The effects of deep learning in medicine have changed the conventional ways of clinical application significantly. Although some sub-fields of medicine, such as pediatrics, have been relatively slow in receiving the critical benefits of deep learning, related research in pediatrics has started to accumulate to a significant level, too. Hence, in this paper, we review recently developed machine learning and deep learning-based solutions for neonatology applications. We systematically evaluate the roles of both classical machine learning and deep learning in neonatology applications, define the methodologies, including algorithmic developments, and describe the remaining challenges in the assessment of neonatal diseases by using PRISMA 2020 guidelines. To date, the primary areas of focus in neonatology regarding AI applications have included survival analysis, neuroimaging, analysis of vital parameters and biosignals, and retinopathy of prematurity diagnosis. We have categorically summarized 106 research articles from 1996 to 2022 and discussed their pros and cons, respectively. In this systematic review, we aimed to further enhance the comprehensiveness of the study. We also discuss possible directions for new AI models and the future of neonatology with the rising power of AI, suggesting roadmaps for the integration of AI into neonatal intensive care units.

Selecting the best optimizers for deep learning-based medical image segmentation.

Purpose: The goal of this work is to explore the best optimizers for deep learning in the context of medical image segmentation and to provide guidance on how to design segmentation networks with effective optimization strategies. Approach: Most successful deep learning networks are trained using two types of stochastic gradient descent (SGD) algorithms: adaptive learning and accelerated schemes. Adaptive learning helps with fast convergence by starting with a larger learning rate (LR) and gradually decreasing it. Momentum optimizers are particularly effective at quickly optimizing neural networks within the accelerated schemes category. By revealing the potential interplay between these two types of algorithms [LR and momentum optimizers or momentum rate (MR) in short], in this article, we explore the two variants of SGD algorithms in a single setting. We suggest using cyclic learning as the base optimizer and integrating optimal values of learning rate and momentum rate. The new optimization function proposed in this work is based on the Nesterov accelerated gradient optimizer, which is more efficient computationally and has better generalization capabilities compared to other adaptive optimizers. Results: We investigated the relationship of LR and MR under an important problem of medical image segmentation of cardiac structures from MRI and CT scans. We conducted experiments using the cardiac imaging dataset from the ACDC challenge of MICCAI 2017, and four different architectures were shown to be successful for cardiac image segmentation problems. Our comprehensive evaluations demonstrated that the proposed optimizer achieved better results (over a 2% improvement in the dice metric) than other optimizers in the deep learning literature with similar or lower computational cost in both single and multi-object segmentation settings. Conclusions: We hypothesized that the combination of accelerated and adaptive optimization methods can have a drastic effect in medical image segmentation performances. To this end, we proposed a new cyclic optimization method (Cyclic Learning/Momentum Rate) to address the efficiency and accuracy problems in deep learning-based medical image segmentation. The proposed strategy yielded better generalization in comparison to adaptive optimizers.

A review of deep learning and radiomics approaches for pancreatic cancer diagnosis from medical imaging.

PURPOSE OF REVIEW: Early and accurate diagnosis of pancreatic cancer is crucial for improving patient outcomes, and artificial intelligence (AI) algorithms have the potential to play a vital role in computer-aided diagnosis of pancreatic cancer. In this review, we aim to provide the latest and relevant advances in AI, specifically deep learning (DL) and radiomics approaches, for pancreatic cancer diagnosis using cross-sectional imaging examinations such as computed tomography (CT) and magnetic resonance imaging (MRI). RECENT FINDINGS: This review highlights the recent developments in DL techniques applied to medical imaging, including convolutional neural networks (CNNs), transformer-based models, and novel deep learning architectures that focus on multitype pancreatic lesions, multiorgan and multitumor segmentation, as well as incorporating auxiliary information. We also discuss advancements in radiomics, such as improved imaging feature extraction, optimized machine learning classifiers and integration with clinical data. Furthermore, we explore implementing AI-based clinical decision support systems for pancreatic cancer diagnosis using medical imaging in practical settings. SUMMARY: Deep learning and radiomics with medical imaging have demonstrated strong potential to improve diagnostic accuracy of pancreatic cancer, facilitate personalized treatment planning, and identify prognostic and predictive biomarkers. However, challenges remain in translating research findings into clinical practice. More studies are required focusing on refining these methods, addressing significant limitations, and developing integrative approaches for data analysis to further advance the field of pancreatic cancer diagnosis.

Radiomics Boosts Deep Learning Model for IPMN Classification.

Intraductal Papillary Mucinous Neoplasm (IPMN) cysts are pre-malignant pancreas lesions, and they can progress into pancreatic cancer. Therefore, detecting and stratifying their risk level is of ultimate importance for effective treatment planning and disease control. However, this is a highly challenging task because of the diverse and irregular shape, texture, and size of the IPMN cysts as well as the pancreas. In this study, we propose a novel computer-aided diagnosis pipeline for IPMN risk classification from multi-contrast MRI scans. Our proposed analysis framework includes an efficient volumetric self-adapting segmentation strategy for pancreas delineation, followed by a newly designed deep learning-based classification scheme with a radiomics-based predictive approach. We test our proposed decision-fusion model in multi-center data sets of 246 multi-contrast MRI scans and obtain superior performance to the state of the art (SOTA) in this field. Our ablation studies demonstrate the significance of both radiomics and deep learning modules for achieving the new SOTA performance compared to international guidelines and published studies (81.9% vs 61.3% in accuracy). Our findings have important implications for clinical decision-making. In a series of rigorous experiments on multi-center data sets (246 MRI scans from five centers), we achieved unprecedented performance (81.9% accuracy). The code is available upon publication.

Microglial iron trafficking: new player in brain injury.

BACKGROUND: Neonatal brain injury is a significant reason of neurodevelopmental abnormalities and long-term neurological impairments. Hypoxic-ischemic encephalopathy and preterm brain injury, including intraventricular hemorrhage are the most common grounds of brain injury for full-term and preterm neonates. The prevalence of hypoxic ischemic encephalopathy varies globally, ranging from 1 to 3.5/1000 live births in high-resource countries and 26/1000 in low-resource countries. Preterm birth's global incidence is 15 million, a significant reason for infant mortality and morbidity, permanent neurologic problems, and the associated social and economic burden. The widespread neurodevelopmental effects of neonatal brain injury could have an unfavorable impact on a variety of aspects of cognitive, linguistic, behavioral, sensory, and motor functions. Brain injury occurs via various mechanisms, including energy deprivation, excitatory amino acids, mitochondrial dysfunction, reactive oxygen species, and inflammation giving rise to different forms of cell death. The contribution of microglial activity in neonatal brain injury has widely been underlined by focusing on cell death mechanisms since the neuronal death pathways during their development are distinct from those in the adult brain. Iron accumulation and lipid peroxidation cause a relatively novel type of regulated cell death called ferroptosis. Neonates generally have biochemical iron inequalities, and their antioxidant potential is highly restricted, implying that ferroptosis may be significant in pathologic conditions. Moreover, inhaled nitric oxide therapy in infants may lead to microglial inflammation via ferroptosis and neuronal injury in the developing brain. This review article aims to summarize the studies that investigated the association between neonatal brain injury and iron metabolism, with a particular emphasis on the microglial activity and its application to the inhibition of neonatal brain injury.

Serum Creatinine Patterns in Neonates Treated with Therapeutic Hypothermia for Neonatal Encephalopathy.

INTRODUCTION: There is large variability in kidney function and injury in neonates with neonatal encephalopathy (NE) treated with therapeutic hypothermia (TH). Acute kidney injury (AKI) definitions that apply categorical approaches may lose valuable information about kidney function in individual patients. Centile serum creatinine (SCr) over postnatal age (PNA) may provide more valuable information in TH neonates. METHODS: Data from seven TH neonates and one non-TH-treated, non-NE control cohorts were pooled in a retrospective study. SCr centiles over PNA, and AKI incidence (definition: SCr ↑≥0.3 mg/dL within 48 h, or ↑ ≥1.5 fold vs. the lowest prior SCr within 7 days) and mortality were calculated. Repeated measurement linear models were applied to SCr trends, modeling SCr on PNA, birth weight or gestational age (GA), using heterogeneous autoregressive residual covariance structure and maximum likelihood methods. Findings were compared to patterns in the control cohort. RESULTS: Among 1,136 TH neonates, representing 4,724 SCr observations, SCr (10th-25th-50th-75th-90th-95th) PNA centiles (day 1-10) were generated. In TH neonates, the AKI incidence was 132/1,136 (11.6%), mortality 193/1,136 (17%). AKI neonates had a higher mortality (37.2-14.3%, p < 0.001). Median SCr patterns over PNA were significantly higher in nonsurvivors (p < 0.01) or AKI neonates (p < 0.001). In TH-treated neonates, PNA and GA or birth weight explained SCr variability. Patterns over PNA were significantly higher in TH neonates to controls (801 neonates, 2,779 SCr). CONCLUSIONS: SCr patterns in TH-treated NE neonates are specific. Knowing PNA-related patterns enable clinicians to better assess kidney function and tailor pharmacotherapy, fluids, or kidney supportive therapies.

COX-1, COX-2 and CYP2C19 variations may be related to cardiovascular events due to acetylsalicylic acid resistance.

BACKGROUND: In some stent implanted patients, cardiovascular events (CE) may occur. Acetylsalicylic acid (ASA) is routinely administered to these patients in order to prevent the occurrence of CE. CE may be related to gene variations which cause ASA resistance (AR). Therefore, it was aimed to investigate the relationship between COX-1, COX-2, CYP2C9 and CYP2C19 variations with CE due to AR. MATERIALS AND RESULTS: Seventy-four stent implanted patients, using 100 mg of ASA per day during five years were enrolled into the study. Following stent implantation, thirty-eight patients who had a CE within five years due to AR and 36 patients without CE were enrolled in patient and control group, respectively. AR was confirmed by platelet aggregation testing. After DNA isolation from blood; COX-1, COX-2, CYP2C19 and CYP2C9 variations were investigated with real-time polymerase chain reaction. At the end of this study, heterozygous genotype of COX-1 was found statistically high in patients whereas heterozygous genotype of CYP2C19*17 was found statistically high in controls. The presence of C and G allele in COX-1 and COX-2 were found statistically high in patients, respectively. The presence of T allele in CYP2C19*17 was found statistically high in controls. Heterozygous genotype of COX-1 variation was found statistically high in patients who have AR. Additionally heterozygous genotype of CYP2C19*17 was found statistically high in patients who have low thrombosis risk. CONCLUSIONS: COX-1 and COX-2 gene mutations may increase the risk of CE due to AR whereas CYP2C19*17 may have a protective effect in this process.

Transformer based Generative Adversarial Network for Liver Segmentation.

Automated liver segmentation from radiology scans (CT, MRI) can improve surgery and therapy planning and follow-up assessment in addition to conventional use for diagnosis and prognosis. Although convolutional neural networks (CNNs) have became the standard image segmentation tasks, more recently this has started to change towards Transformers based architectures because Transformers are taking advantage of capturing long range dependence modeling capability in signals, so called attention mechanism. In this study, we propose a new segmentation approach using a hybrid approach combining the Transformer(s) with the Generative Adversarial Network (GAN) approach. The premise behind this choice is that the self-attention mechanism of the Transformers allows the network to aggregate the high dimensional feature and provide global information modeling. This mechanism provides better segmentation performance compared with traditional methods. Furthermore, we encode this generator into the GAN based architecture so that the discriminator network in the GAN can classify the credibility of the generated segmentation masks compared with the real masks coming from human (expert) annotations. This allows us to extract the high dimensional topology information in the mask for biomedical image segmentation and provide more reliable segmentation results. Our model achieved a high dice coefficient of 0.9433, recall of 0.9515, and precision of 0.9376 and outperformed other Transformer based approaches. The implementation details of the proposed architecture can be found at https://github.com/UgurDemir/tranformer_liver_segmentation.

Information Bottleneck Attribution for Visual Explanations of Diagnosis and Prognosis.

Visual explanation methods have an important role in the prognosis of the patients where the annotated data is limited or unavailable. There have been several attempts to use gradient-based attribution methods to localize pathology from medical scans without using segmentation labels. This research direction has been impeded by the lack of robustness and reliability. These methods are highly sensitive to the network parameters. In this study, we introduce a robust visual explanation method to address this problem for medical applications. We provide an innovative visual explanation algorithm for general purpose and as an example application we demonstrate its effectiveness for quantifying lesions in the lungs caused by the Covid-19 with high accuracy and robustness without using dense segmentation labels. This approach overcomes the drawbacks of commonly used Grad-CAM and its extended versions. The premise behind our proposed strategy is that the information flow is minimized while ensuring the classifier prediction stays similar. Our findings indicate that the bottleneck condition provides a more stable severity estimation than the similar attribution methods. The source code will be publicly available upon publication.

Successful application of CytoSorb® hemadsorption in an immunocompromised teenager with collapsing glomerulopathy, acute respiratory distress syndrome, and sepsis.

OBJECTIVE: CytoSorb® hemadsorption is an adjunctive therapy in order to reduce elevated cytokine levels of interleukin-6, interleukin-1, and tumor necrosis factor alpha. Here we present a successful administration of CytoSorb® hemadsorption in an immunocompromised pediatric patient with collapsing glomerulopathy, acute respiratory distress syndrome, and sepsis. DATA SOURCES: Clinical observations of one patient. STUDY SELECTION: Case report. DATA EXTRACTION: Data sources are clinical observation during patient management and patient's medical records if needed. The patient's consent was obtained prior to the study. DATA SYNTHESIS: A 17-year-old male with diarrhea was admitted to the hospital and was later found to have elevated creatinine levels and proteinuria. The renal biopsy was consistent with collapsing glomerulopathy and treatment with multi immunosuppressive agents including corticosteroids, mycophenolate mofetil, and rituximab coupled with several courses of hemodialysis and plasmapheresis were administered. During the hospital stay, Stenotrophomonas maltophilia bacteremia from the blood and the catheter cultures were identified. No clinical response was achieved, and patient developed severe sepsis despite antibiotics, intravenous immunoglobulin, and supportive management including albumin, platelet and erythrocyte concentrations, and fresh frozen plasma. CytoSorb® hemadsorption was then added to the ongoing treatment for three consecutive days. Subsequent to CytoSorb® hemadsorption, immediate laboratory and clinical response were observed. CONCLUSION: This is the successful clinical report of an immunocompromised teenager with collapsing nephropathy, sepsis, and multi-organ dysfunction syndrome treated with a combination of renal replacement therapy and CytoSorb® hemadsorption. The usage of CytoSorb® hemadsorption represents a novel approach to improve survival of the patients with multiple organ dysfunction and sepsis.

Early initiated feeding versus early reached target enteral nutrition in critically ill children: An observational study in paediatric intensive care units in Turkey.

AIM: Although early enteral nutrition (EN) is strongly associated with lower mortality in critically ill children, there is no consensus on the definition of early EN. The aim of this study was to evaluate our current practice supplying EN and to identify factors that affect both the initiation of feeding within 24 h after paediatric intensive care unit (PICU) admission and the adequate supply of EN in the first 48 h after PICU admission in critically ill children. METHODS: We conducted a prospective, multicentre, observational study in nine PICUs in Turkey. Any kind of tube feeding commenced within 24 h of PICU admission was considered early initiated feeding (EIF). Patients who received more than 25% of the estimated energy requirement via enteral feeding within 48 h of PICU admission were considered to have early reached target EN (ERTEN). RESULTS: Feeding was initiated in 47.4% of patients within 24 h after PICU admission. In many patients, initiation of feeding seems to have been delayed without an evidence-based reason. ERTEN was achieved in 43 (45.3%) of 95 patients. Patients with EIF were significantly more likely to reach ERTEN. ERTEN was an independent significant predictor of mortality (P < 0.001), along with reached target enteral caloric intake on day 2 associated with decreased mortality. CONCLUSIONS: There is a substantial variability among clinicians' perceptions regarding indications for delay to initiate enteral feeding in critically ill children, especially after the first 6 h of PICU admission. ERTEN, but not EIF, is associated with a significantly lower mortality rate in critically ill children.

Synthesis of antibacterial amphiphilic elastomer based on polystyrene-block-polyisoprene-block-polystyrene via thiol-ene addition.

A new type of amphiphilic antibacterial elastomer has been described. Thermoplastic elastomer, polystyrene-block-polyisoprene-block-polystyrene (PS-b-PI-b-PS) triblock copolymer was functionalized in toluene solution by free radical mercaptan addition in order to obtain an amphiphilic antibacterial elastomer. Thiol terminated PEG was grafted through the double bonds of PS-b-PI-b-PS via free radical thiol-ene coupling reaction. The antibacterial properties of the amphiphilic graft copolymers were observed. The original and the modified polymers were used to create microfibers in an electro-spinning process. Topology of the electrospun micro/nanofibers were studied by using scanning electron microscopy (SEM). The chemical structures of the amphiphilic comb type graft copolymers were elucidated by the combination of elemental analysis, (1)H NMR, (13)C NMR, GPC and FTIR.

Apicidin suppresses transcription of 17ß-hydroxysteroid dehydrogenase type 1 in endometrial adenocarcinoma cells.

It has recently been reported that endometrial cancer cells are able to convert estron (E1) to 17ß estradiol (E2). We observed the presence of 17ß-hydroxysteroid dehydrogenase type 1 (HSD17B1) transcript and protein in receptor positive ER(+) and negative ER(-) Ishikawa endometrial adenocarcinoma (ISH) cells. ER(+) ISH, but not ER(-)02 ISH, cells were significantly susceptible to apicidin induced death, and we further used ER(-)ISH cells to study the effect of apicidin on cellular levels of HSD17B1 transcript and protein. We showed that apicidin significantly lowered HSD17B1 transcript and protein levels in ISH cells. There was no significant effect on HSD17B1 transcript stability. However, chromatin immunoprecipitation analysis revealed that apicidin significantly decreased occupation of the first exon of the HSD17B1 gene by Polymerase II. Since intratumoral E1 to E2 conversion is a significant contributor to the progression of estrogen dependent cancers, and HDAC inhibitors are being tested in anticancer clinical trials, our observations may have clinical value.

The distribution of the intestinal parasitic diseases in the Southeast Anatolian (GAP=SEAP) region of Turkey.

OBJECTIVES: The physical alterations put in place by the Southeastern Anatolia Project will undoubtedly provide a remarkable economical growth and a social development in the area. In addition, the influence that formation of dam ponds, enlargement of irrigation areas, change of product and the way of cultivation, urbanization and industrialization will have an impact on the environment. To minimize the adverse effects of this process on human beings, a Community Health Project was completed by the teams participated by Ege, Dicle, Gaziantep and Harran Universities under the Directorate of Turkish Parasitology Association and by Southeastern Anatolia Project Regional Development Administration between 2001 and 2003. RESULTS: To identify individuals with parasite, feces samples were taken from a total of 4,470 individuals. Parasites were found in feces of 41.8% of men, 44.3% of women and 32.2% of children, 0-59 months old, who were included in the research and gave feces samples for parasites tests. These prevalence values indicate how widespread parasitic diseases are in the region. The high prevalence of parasitic diseases in this area is one of the causes of malnutrition in 40% of children. Parasites were detected in 44.2% of feces samples taken from rural areas and in 39.5% taken from urban areas. When the distribution of parasites detected in feces samples was studied, the most common parasites were Giardia intestinalis (18.1%), Entamoeba coli (11.8%), Ascaris lumbricoides (4.8%), Trichuris trichiura (4.5%) and Hymenolepis nana (3.9%). Distribution of parasites according to cities varied widely. The most frequently seen parasites were T. trichiura in Gaziantep; G. intestinalis in Batman, Mardin, Diyarbakir, Sirnak and Sanliurfa; and E. coli in Siirt, Kilis and Adiyaman. CONCLUSIONS: This study is the first investigation of intestinal parasite prevalence in a large region, specifically, in this GAP region and in Turkey, in general. There is no direct relationship between irrigating the cultivation areas and diffusion of parasitic diseases because the existence of intestinal parasites mentioned above is not related to the range of irrigation of cultivation areas, but is related to factors already discussed.