Multi-Dimensional Medical Image Fusion With Complex Sparse Representation.

In the field of medical imaging, the fusion of data from diverse modalities plays a pivotal role in advancing our understanding of pathological conditions. Sparse representation (SR), a robust signal modeling technique, has demonstrated noteworthy success in multi-dimensional (MD) medical image fusion. However, a fundamental limitation appearing in existing SR models is their lack of directionality, restricting their efficacy in extracting anatomical details from different imaging modalities. To tackle this issue, we propose a novel directional SR model, termed complex sparse representation (ComSR), specifically designed for medical image fusion. ComSR independently represents MD signals over directional dictionaries along specific directions, allowing precise analysis of intricate details of MD signals. Besides, current studies in medical image fusion mostly concentrate on addressing either 2D or 3D fusion problems. This work bridges this gap by proposing a MD medical image fusion method based on ComSR, presenting a unified framework for both 2D and 3D fusion tasks. Experimental results across six multi-modal medical image fusion tasks, involving 93 pairs of 2D source images and 20 pairs of 3D source images, substantiate the superiority of our proposed method over 11 state-of-the-art 2D fusion methods and 4 representative 3D fusion methods, in terms of both visual quality and objective evaluation. The source code of our fusion method is available at https://github.com/Imagefusions/imagefusions/tree/main.

A Combined Model Integrating Radiomics and Deep Learning Based on Contrast-Enhanced CT for Preoperative Staging of Laryngeal Carcinoma.

RATIONALE AND OBJECTIVES: Accurate staging of laryngeal carcinoma can inform appropriate treatment decision-making. We developed a radiomics model, a deep learning (DL) model, and a combined model (incorporating radiomics features and DL features) based on the venous-phase CT images and explored the performance of these models in stratifying patients with laryngeal carcinoma into stage I-II and stage III-IV, and also compared these models with radiologists. MATERIALS AND METHODS: Three hundreds and nineteen patients with pathologically confirmed laryngeal carcinoma were randomly divided into a training set (n = 223) and a test set (n = 96). In the training set, the radiomics features with inter- and intraclass correlation coefficients (ICCs)> 0.75 were screened by Spearman correlation analysis and recursive feature elimination (RFE); then support vector machine (SVM) classifier was applied to develop the radiomics model. The DL model was built using ResNet 18 by the cropped 2D regions of interest (ROIs) in the maximum tumor ROI slices and the last fully connected layer of this network served as the DL feature extractor. Finally, a combined model was developed by pooling the radiomics features and extracted DL features to predict the staging. RESULTS: The area under the curves (AUCs) for radiomics model, DL model, and combined model in the test set were 0.704 (95% confidence interval [CI]: 0.588-0.820), 0.724 (95% CI: 0.613-0.835), and 0.849 (95% CI: 0.755-0.943), respectively. The combined model outperformed the radiomics model and the DL model in discriminating stage I-II from stage III-IV (p = 0.031 and p = 0.020, respectively). Only the combined model performed significantly better than radiologists (p < 0.050 for both). CONCLUSION: The combined model can help tailor the therapeutic strategy for laryngeal carcinoma patients by enabling more accurate preoperative staging.

Using deeply time-series semantics to assess depressive symptoms based on clinical interview speech.

Introduction: Depression is an affective disorder that contributes to a significant global burden of disease. Measurement-Based Care (MBC) is advocated during the full course management, with symptom assessment being an important component. Rating scales are widely used as convenient and powerful assessment tool, but they are influenced by the subjectivity and consistency of the raters. The assessment of depressive symptoms is usually conducted with a clear purpose and restricted content, such as clinical interviews based on the Hamilton Depression Rating Scale (HAMD), so that the results are easy to obtain and quantify. Artificial Intelligence (AI) techniques are used due to their objective, stable and consistent performance, and are suitable for assessing depressive symptoms. Therefore, this study applied Deep Learning (DL)-based Natural Language Processing (NLP) techniques to assess depressive symptoms during clinical interviews; thus, we proposed an algorithm model, explored the feasibility of the techniques, and evaluated their performance. Methods: The study included 329 patients with Major Depressive Episode. Clinical interviews based on the HAMD-17 were conducted by trained psychiatrists, whose speech was simultaneously recorded. A total of 387 audio recordings were included in the final analysis. A deeply time-series semantics model for the assessment of depressive symptoms based on multi-granularity and multi-task joint training (MGMT) is proposed. Results: The performance of MGMT is acceptable for assessing depressive symptoms with an F1 score (a metric of model performance, the harmonic mean of precision and recall) of 0.719 in classifying the four-level severity of depression and an F1 score of 0.890 in identifying the presence of depressive symptoms. Disscussion: This study demonstrates the feasibility of the DL and the NLP techniques applied to the clinical interview and the assessment of depressive symptoms. However, there are limitations to this study, including the lack of adequate samples, and the fact that using speech content alone to assess depressive symptoms loses the information gained through observation. A multi-dimensional model combing semantics with speech voice, facial expression, and other valuable information, as well as taking into account personalized information, is a possible direction in the future.

Acute ischemic Stroke combined with Stanford type A aortic dissection: A case report and literature review.

BACKGROUND: Acute aortic dissection (AAD) is a high mortality disease that can lead to acute ischemic strokes (AIS). Some of the patients with AAD combined with AIS initially present with neurological symptoms, which can easily lead to missed or delayed AAD diagnosis. This is attributed to the lack of physician awareness or the urgency of patient thrombolysis. Intravenous administration of thrombolytic therapy (IVT) for AAD is associated with poor prognostic outcomes. We report a patient with AIS combined with AAD who developed a massive cerebral infarction after receiving IVT for a missed AAD diagnosis. CASE SUMMARY: A 49-year-old man was admitted to a local hospital with an acute onset of left-sided limb weakness accompanied by slurred speech. The patient had a history of hypertension that was not regularly treated with medication. Physical examination revealed incomplete mixed aphasia and left limb hemiparesis. Cranial computed tomography (CT) scan showed bilateral basal ganglia and lateral ventricular paraventricular infarct lesions. The patient was diagnosed with AIS and was administered with IVT. After IVT, patient's muscle strength and consciousness deteriorated. From the local hospital, he was referred to our hospital for further treatment. Emergency head and neck CT angiography (CTA) scans were performed. Results showed multiple cerebral infarctions, and aortic dissection in the ascending aorta, innominate artery, as well as in the right common carotid artery. Then, the CTA of thoracoabdominal aorta was performed, which revealed a Stanford type A aortic dissection and aortic dissection extending from the aortic root to the left external iliac artery. Laceration was located in the lesser curvature of the aortic arch. AAD complicated with AIS was considered, and the patient was immediately subjected to cardiovascular surgery for treatment. The next day, the patient underwent aortic arch and ascending aortic replacement and aortic valvuloplasty. CONCLUSION: Clinical manifestations for AAD combined with AIS are diverse. Some patients may not exhibit typical chest or back pains. Therefore, patients should be carefully evaluated to exclude AAD before administering IVT in order to avoid adverse consequences.

Predicting the effective thermal conductivity of unfrozen soils with various water contents based on artificial neural network.

The effective thermal conductivity of soils is a crucial parameter for many applications such as geothermal engineering, environmental science, and agriculture and engineering. However, it is pretty challenging to accurately determine it due to soils' complex structure and components. In the present study, the influences of different parameters, including silt content (msi), sand content (msa), clay content (mcl), quartz content (mqu), porosity, and water content on the effective thermal conductivity of soils, were firstly analyzed by the Pearson correlation coefficient. Then different artificial neural network (ANN) models were developed based on the 465 groups of thermal conductivity of unfrozen soils collected from the literature to predict the effective thermal conductivity of soils. Results reveal that the parameters ofmsi,msa,mcl, andmquhave a relatively slight influence on the effective thermal conductivity of soils compared to the water content and porosity. Although the ANN model with six parameters has the highest accuracy, the ANN model with two input parameters (porosity and water content) could predict the effective thermal conductivity well with acceptable accuracy andR2= 0.940. Finally, a correlation of the effective thermal conductivity for different soils was proposed based on the large number of results predicted by the two input parameters ANN-based model. This correlation has proved to have a higher accuracy without assumptions and uncertain parameters when compared to several commonly used existing models.

MicroRNAs as potential biomarkers for the diagnosis of Traumatic Brain Injury: A systematic review and meta-analysis.

Background: Traumatic brain injury (TBI) is a sudden trauma on the head, in which severe TBI (sTBI) is usually associated with death and long-term disability. MicroRNAs (miRNAs) are potential biomarkers of diverse diseases, including TBI. However, few systematic reviews and meta-analyses have been conducted to determine the clinical value of miRNAs expression in TBI patients. Methods: We conducted this systematic review and meta-analysis study according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We searched PubMed, Embase, the Cochrane Library, Web of Science, from inception to August 26, 2020. We included articles written in English that have reported on the diagnostic value of miRNAs expression in TBI patients. We excluded studies that did not provided sufficient information to construct the 2×2 contingency table. Results: Eight studies investigating the diagnostic value of miRNA in TBI were analyzed in this study. The overall sensitivity, specificity and area under the curve (AUC) of miRNAs in diagnosis of TBI were 89% [95% confidence interval (CI): 0.84-0.93], 92% (95% CI 0.82-0.97) and 95% (95% CI 0.93-0.97). We found that panels of multiple miRNAs could improve the diagnostic accuracy of TBI. Samples from blood and brain tissue have significantly enhanced diagnostic accuracy, when compared with saliva. The AUC of miRNAs in severe TBI was 0.97, with 91% sensitivity and 92% specificity. Conclusion: This systematic review and meta-analysis demonstrated that miRNAs could be potential diagnostic markers in TBI patients. MiRNAs detected in blood and brain tissue display high accuracy for TBI diagnosis.

Master clinical medical knowledge at certificated-doctor-level with deep learning model.

Mastering of medical knowledge to human is a lengthy process that typically involves several years of school study and residency training. Recently, deep learning algorithms have shown potential in solving medical problems. Here we demonstrate mastering clinical medical knowledge at certificated-doctor-level via a deep learning framework Med3R, which utilizes a human-like learning and reasoning process. Med3R becomes the first AI system that has successfully passed the written test of National Medical Licensing Examination in China 2017 with 456 scores, surpassing 96.3% human examinees. Med3R is further applied for providing aided clinical diagnosis service based on real electronic medical records. Compared to human experts and competitive baselines, our system can provide more accurate and consistent clinical diagnosis results. Med3R provides a potential possibility to alleviate the severe shortage of qualified doctors in countries and small cities of China by providing computer-aided medical care and health services for patients.

Emulsified isoflurane treatment inhibits the cell cycle and respiration of human bronchial epithelial 16HBE cells in a p53-independent manner.

Emulsified isoflurane (EIso), as a result of its rapid anesthetic induction, recovery and convenience, is widely used as a novel intravenous general anesthetic. Treatment with EIso can reduce injuries caused by ischemia/reperfusion (I/R) to organs, including the heart, lung and liver, without knowing understanding the molecular mechanism. The present study hypothesized that treatment with EIso can affect the physiological processes of human lung bronchial epithelial cells (16HBE) prior to I/R. To test this hypothesis, the present study first constructed stable p53 knockdown and synthesis of cytochrome c oxidase (SCO)2 knockdown 16HBE cells. The above cells were subsequently treated with EIso at a concentration of 0.1 and 0.2% for 24 h. The relevant concentration of fat emulsion was used as a negative control. The expression levels of p53, p21, SCO1, SCO2 and Tp53­induced glycolysis and apoptosis regulator (TIGAR) were detected by reverse transcription­quantitative polymerase chain reaction and western blotting. Subsequently, the cell proliferation, respiration and glycolysis were investigated. The results revealed that EIso treatment significantly decreased the transcription of TIGAR, SCO1 and SCO2, and increased the transcription of p21, which are all p53 target genes, in a p53-independent manner. The cell cycle was inhibited by arresting cells at the G0/G1 phase. Respiration was reduced, which caused a decrease in oxygen consumption and the accumulation of lactate and reactive oxygen species. Taken together, EIso treatment inhibited the proliferation and respiration, and promoted glycolysis in 16HBE cells. This regulatory pathway may represent a protective mechanism of EIso treatment by inhibiting cell growth and decreasing the oxygen consumption from I/R.

Tailored Electrospinning of WO3 Nanobelts as Efficient Ultraviolet Photodetectors with Photo-Dark Current Ratios up to 1000.

In this work, polycrystalline WO3 nanobelts were fabricated via an electrospinning process combined with subsequent air calcination. The resultant products were characterized by X-ray diffraction, field-emission scanning electron microscopy, and high-resolution transmission electron microscopy in regard to the structures. It has been found that the applied voltage during the electrospinning process played the determined role in the formation of the WO3 nanobelts, allowing the controlled growth of the nanobelts. The ultraviolet (UV) photodetector assembled by an individual WO3 nanobelt exhibits a high sensitivity and a precise selectivity to the different wavelength lights, with a very low dark current and typical photo-dark current ratio up to 1000, which was the highest for any WO3 photodectectors ever reported. This work could not only push forward the facile preparation of WO3 nanobelts but also represent, for the first time, the possibility that the polycrystalline WO3 nanobelts could be a promising building block for the highly efficient UV photodetectors.