Exploring a frequency-domain attention-guided cascade U-Net: Towards spatially tunable segmentation of vasculature

Developing fully automatic and highly accurate medical image segmentation methods is critically important for vascular disease diagnosis and treatment planning. Although advances in convolutional neural networks (CNNs) have spawned an array of automatic segmentation models converging to saturated high performance, none have explored whether CNNs can achieve (spatially) tunable segmentation. As a result, we propose multiple attention modules from a frequency-domain perspective to construct a unified CNN architecture for segmenting vasculature with desired (spatial) scales. The proposed CNN architecture is named frequency-domain attention-guided cascaded U-Net (FACU-Net). Specifically, FACU-Net contains two innovative components: (1) a frequency-domain-based channel attention module that adaptively tunes channel-wise feature responses and (2) a frequency-domain-based spatial attention module that enables the deep network to concentrate on foreground regions of interest (ROIs) effectively. Furthermore, we devised a novel frequency-domain-based content attention module to enhance or weaken the high (spatial) frequency information, allowing us to strengthen or eliminate vessels of interest. Extensive experiments using clinical data from patients with intracranial aneurysms (IA) and abdominal aortic aneurysms (AAA) demonstrated that the proposed FACU-Net met its design goal. In addition, we further investigated the association between varying (spatial) frequency components and the desirable vessel size/scale attributes. In summary, our preliminary findings are encouraging, and further developments may lead to deployable image segmentation models that are spatially tunable for clinical applications.

S-Net: a multiple cross aggregation convolutional architecture for automatic segmentation of small/thin structures for cardiovascular applications

With the success of U-Net or its variants in automatic medical image segmentation, building a fully convolutional network (FCN) based on an encoder-decoder structure has become an effective end-to-end learning approach. However, the intrinsic property of FCNs is that as the encoder deepens, higher-level features are learned, and the receptive field size of the network increases, which results in unsatisfactory performance for detecting low-level small/thin structures such as atrial walls and small arteries. To address this issue, we propose to keep the different encoding layer features at their original sizes to constrain the receptive field from increasing as the network goes deeper. Accordingly, we develop a novel S-shaped multiple cross-aggregation segmentation architecture named S-Net, which has two branches in the encoding stage, i.e., a resampling branch to capture low-level fine-grained details and thin/small structures and a downsampling branch to learn high-level discriminative knowledge. In particular, these two branches learn complementary features by residual cross-aggregation; the fusion of the complementary features from different decoding layers can be effectively accomplished through lateral connections. Meanwhile, we perform supervised prediction at all decoding layers to incorporate coarse-level features with high semantic meaning and fine-level features with high localization capability to detect multi-scale structures, especially for small/thin volumes fully. To validate the effectiveness of our S-Net, we conducted extensive experiments on the segmentation of cardiac wall and intracranial aneurysm (IA) vasculature, and quantitative and qualitative evaluations demonstrated the superior performance of our method for predicting small/thin structures in medical images.

Augmenting Prediction of Intracranial Aneurysms' Risk Status Using Velocity-Informatics: Initial Experience

Our primary goal here is to demonstrate that innovative analytics of aneurismal velocities, named velocity-informatics, enhances intracranial aneurysm (IA) rupture status prediction. 3D computer models were generated using imaging data from 112 subjects harboring anterior IAs (4-25 mm; 44 ruptured and 68 unruptured). Computational fluid dynamics simulations and geometrical analyses were performed. Then, computed 3D velocity vector fields within the IA dome were processed for velocity-informatics. Four machine learning methods (support vector machine, random forest, generalized linear model, and GLM with Lasso or elastic net regularization) were employed to assess the merits of the proposed velocity-informatics. All 4 ML methods consistently showed that, with velocity-informatics metrics, the area under the curve and prediction accuracy both improved by approximately 0.03. Overall, with velocity-informatics, the support vector machine's prediction was most promising: an AUC of 0.86 and total accuracy of 77%, with 60% and 88% of ruptured and unruptured IAs being correctly identified, respectively.

Radiomic-based Textural Analysis of Intraluminal Thrombus in Aortic Abdominal Aneurysms: A Demonstration of Automated Workflow

Our main objective is to investigate how the structural information of intraluminal thrombus (ILT) can be used to predict abdominal aortic aneurysms (AAA) growth status through an automated workflow. Fifty-four human subjects with ILT in their AAAs were identified from our database; those AAAs were categorized as slowly- (< 5 mm/year) or fast-growing (≥ 5 mm/year) AAAs. In-house deep-learning image segmentation models were used to generate 3D geometrical AAA models, followed by automated analysis. All features were fed into a support vector machine classifier to predict AAA's growth status.The most accurate prediction model was achieved through four geometrical parameters measuring the extent of ILT, two parameters quantifying the constitution of ILT, antihypertensive medication, and the presence of co-existing coronary artery disease. The predictive model achieved an AUROC of 0.89 and a total accuracy of 83%. When ILT was not considered, our prediction's AUROC decreased to 0.75 (P-value < 0.001).

Deep-learning-based image segmentation for image-based computational hemodynamic analysis of abdominal aortic aneurysms: a comparison study

Computational hemodynamics is increasingly being used to quantify hemodynamic characteristics in and around abdominal aortic aneurysms (AAA) in a patient-specific fashion. However, the time-consuming manual annotation hinders the clinical translation of computational hemodynamic analysis. Thus, we investigate the feasibility of using deep-learning-based image segmentation methods to reduce the time required for manual segmentation. Two of the latest deep-learning-based image segmentation methods, ARU-Net and CACU-Net, were used to test the feasibility of automated computer model creation for computational hemodynamic analysis. Morphological features and hemodynamic metrics of 30 computed tomography angiography (CTA) scans were compared between pre-dictions and manual models. The DICE score for both networks was 0.916, and the correlation value was above 0.95, indicating their ability to generate models comparable to human segmentation. The Bland-Altman analysis shows a good agreement between deep learning and manual segmentation results. Compared with manual (computational hemodynamics) model recreation, the time for automated computer model generation was significantly reduced (from ∼2 h to ∼10 min). Automated image segmentation can significantly reduce time expenses on the recreation of patient-specific AAA models. Moreover, our study showed that both CACU-Net and ARU-Net could accomplish AAA segmentation, and CACU-Net outperformed ARU-Net in terms of accuracy and time-saving.

Characterization of small abdominal aortic aneurysms' growth status using spatial pattern analysis of aneurismal hemodynamics

Aneurysm hemodynamics is known for its crucial role in the natural history of abdominal aortic aneurysms (AAA). However, there is a lack of well-developed quantitative assessments for disturbed aneurysmal flow. Therefore, we aimed to develop innovative metrics for quantifying disturbed aneurysm hemodynamics and evaluate their effectiveness in predicting the growth status of AAAs, specifically distinguishing between fast-growing and slowly-growing aneurysms. The growth status of aneurysms was classified as fast (≥ 5 mm/year) or slow (< 5 mm/year) based on serial imaging over time. We conducted computational fluid dynamics (CFD) simulations on 70 patients with computed tomography (CT) angiography findings. By converting hemodynamics data (wall shear stress and velocity) located on unstructured meshes into image-like data, we enabled spatial pattern analysis using Radiomics methods, referred to as "Hemodynamics-informatics" (i.e., using informatics techniques to analyze hemodynamic data). Our best model achieved an AUROC of 0.93 and an accuracy of 87.83%, correctly identifying 82.00% of fast-growing and 90.75% of slowly-growing AAAs. Compared with six classification methods, the models incorporating hemodynamics-informatics exhibited an average improvement of 8.40% in AUROC and 7.95% in total accuracy. These preliminary results indicate that hemodynamics-informatics correlates with AAAs' growth status and aids in assessing their progression.

Computerized Differentiation of Growth Status for Abdominal Aortic Aneurysms: A Feasibility Study

Fast-growing abdominal aortic aneurysms (AAA) have a high rupture risk and poor outcomes if not promptly identified and treated. Our primary objective is to improve the differentiation of small AAAs' growth status (fast versus slow-growing) through a combination of patient health information, computational hemodynamics, geometric analysis, and artificial intelligence. 3D computed tomography angiography (CTA) data available for 70 patients diagnosed with AAAs with known growth status were used to conduct geometric and hemodynamic analyses. Differences among ten metrics (out of ninety metrics) were statistically significant discriminators between fast and slow-growing groups. Using a support vector machine (SVM) classifier, the area under receiving operating curve (AUROC) and total accuracy of our best predictive model for differentiation of AAAs' growth status were 0.86 and 77.50%, respectively. In summary, the proposed analytics has the potential to differentiate fast from slow-growing AAAs, helping guide resource allocation for the management of patients with AAAs.

Automatic segmentation of abdominal aortic aneurysms from CT angiography using a context-aware cascaded U-Net

We delineate abdominal aortic aneurysms, including lumen and intraluminal thrombosis (ILT), from contrast-enhanced computed tomography angiography (CTA) data in 70 patients with complete automation. A novel context-aware cascaded U-Net configuration enables automated image segmentation. Notably, auto-context structure, in conjunction with dilated convolutions, anisotropic context module, hierarchical supervision, and a multi-class loss function, are proposed to improve the delineation of ILT in an unbalanced, low-contrast multi-class labeling problem. A quantitative analysis shows that the automated image segmentation produces comparable results with trained human users (e.g., DICE scores of 0.945 and 0.804 for lumen and ILT, respectively). Resultant morphological metrics (e.g., volume, surface area, etc.) are highly correlated to those parameters generated by trained human users. In conclusion, the proposed automated multi-class image segmentation tool has the potential to be further developed as a translational software tool that can be used to improve the clinical management of AAAs.

USING CONVOLUTIONAL NEURAL NETWORK-BASED SEGMENTATION FOR IMAGE-BASED COMPUTATIONAL FLUID DYNAMICS SIMULATIONS OF BRAIN ANEURYSMS: INITIAL EXPERIENCE IN AUTOMATED MODEL CREATION

"Image-based" computational fluid dynamics (CFD) simulations provide insights into each patient's hemodynamic environment. However, current standard procedures for creating CFD models start with manual segmentation and are time-consuming, hindering the clinical translation of image-based CFD simulations. This feasibility study adopts deep-learning-based image segmentation (hereafter referred to as Artificial Intelligence (AI) segmentation) to replace manual segmentation to accelerate CFD model creation. Two published convolutional neural network-based AI methods (MIScnn and DeepMedic) were selected to perform CFD model extraction from three-dimensional (3D) rotational angiography data containing intracranial aneurysms. In this study, aneurysm morphological and hemodynamic results using models generated by AI segmentation methods were compared with those obtained by two human users for the same data. Interclass coefficients (ICC), Bland-Altman plots, and Pearson's correlation coefficients (PCC) were combined to assess how well AI-generated CFD models were performed. We found that almost perfect agreement was obtained between the human and AI results for all eleven morphological and five out of eight hemodynamic parameters, while a moderate agreement was obtained from the remaining three hemodynamic parameters. Given this level of agreement, using AI segmentation to create CFD models is feasible, given more developments.

Can we explain machine learning-based prediction for rupture status assessments of intracranial aneurysms?

Although applying machine learning (ML) algorithms to rupture status assessment of intracranial aneurysms (IA) has yielded promising results, the opaqueness of some ML methods has limited their clinical translation. We presented the first explainability comparison of six commonly used ML algorithms: multivariate logistic regression (LR), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGBoost), multi-layer perceptron neural network (MLPNN), and Bayesian additive regression trees (BART). A total of 112 IAs with known rupture status were selected for this study. The ML-based classification used two anatomical features, nine hemodynamic parameters, and thirteen morphologic variables. We utilized permutation feature importance, local interpretable model-agnostic explanations (LIME), and SHapley Additive exPlanations (SHAP) algorithms to explain and analyze 6 Ml algorithms. All models performed comparably: LR area under the curve (AUC) was 0.71; SVM AUC was 0.76; RF AUC was 0.73; XGBoost AUC was 0.78; MLPNN AUC was 0.73; BART AUC was 0.73. Our interpretability analysis demonstrated consistent results across all the methods; i.e., the utility of the top 12 features was broadly consistent. Furthermore, contributions of 9 important features (aneurysm area, aneurysm location, aneurysm type, wall shear stress maximum during systole, ostium area, the size ratio between aneurysm width, (parent) vessel diameter, one standard deviation among time-averaged low shear area, and one standard deviation of temporally averaged low shear area less than 0.4 Pa) were nearly the same. This research suggested that ML classifiers can provide explainable predictions consistent with general domain knowledge concerning IA rupture. With the improved understanding of ML algorithms, clinicians' trust in ML algorithms will be enhanced, accelerating their clinical translation.

An attention residual u-net with differential preprocessing and geometric postprocessing: Learning how to segment vasculature including intracranial aneurysms

OBJECTIVE

Intracranial aneurysms (IA) are lethal, with high morbidity and mortality rates. Reliable, rapid, and accurate segmentation of IAs and their adjacent vasculature from medical imaging data is important to improve the clinical management of patients with IAs. However, due to the blurred boundaries and complex structure of IAs and overlapping with brain tissue or other cerebral arteries, image segmentation of IAs remains challenging. This study aimed to develop an attention residual U-Net (ARU-Net) architecture with differential preprocessing and geometric postprocessing for automatic segmentation of IAs and their adjacent arteries in conjunction with 3D rotational angiography (3DRA) images.

METHODS

The proposed ARU-Net followed the classic U-Net framework with the following key enhancements. First, we preprocessed the 3DRA images based on boundary enhancement to capture more contour information and enhance the presence of small vessels. Second, we introduced the long skip connections of the attention gate at each layer of the fully convolutional decoder-encoder structure to emphasize the field of view (FOV) for IAs. Third, residual-based short skip connections were also embedded in each layer to implement in-depth supervision to help the network converge. Fourth, we devised a multiscale supervision strategy for independent prediction at different levels of the decoding path, integrating multiscale semantic information to facilitate the segmentation of small vessels. Fifth, the 3D conditional random field (3DCRF) and 3D connected component optimization (3DCCO) were exploited as postprocessing to optimize the segmentation results.

RESULTS

Comprehensive experimental assessments validated the effectiveness of our ARU-Net. The proposed ARU-Net model achieved comparable or superior performance to the state-of-the-art methods through quantitative and qualitative evaluations. Notably, we found that ARU-Net improved the identification of arteries connecting to an IA, including small arteries that were hard to recognize by other methods. Consequently, IA geometries segmented by the proposed ARU-Net model yielded superior performance during subsequent computational hemodynamic studies (also known as "patient-specific" computational fluid dynamics [CFD] simulations). Furthermore, in an ablation study, the five key enhancements mentioned above were confirmed.

CONCLUSIONS

The proposed ARU-Net model can automatically segment the IAs in 3DRA images with relatively high accuracy and potentially has significant value for clinical computational hemodynamic analysis.

[Bilateral superior cervical ganglionectomy attenuates cardiac remodeling and improves cardiac function in pressure-overloaded heart failure mice]

Objective: To investigate the effect of bilateral superior cervical ganglionectomy on cardiac remodeling and function in pressure-overloaded heart failure (HF) mice. Methods: Pressure-overloaded HF mouse model was produced by severe thoracic aorta banding (sTAB). Bilateral superior cervical ganglionectomy (SCGx) was performed 2 weeks after sTAB. Twenty four 6-week-old male C57BL/6 mice were randomized divided into 4 groups (n=6 each): control group: sham sTAB+sham SCGx; denervated group: sham sTAB+SCGx; HF group: sTAB+sham SCGx; denervated HF group: sTAB+SCGx. Cardiac function was measured by echocardiography at week 0, 1, 2, and 4 after sTAB, respectively. All mice were sacrificed at the end of week 4 and heart tissues were harvested. HE and Masson staining were performed. Immunohistochemical staining (IHC) for tyrosine hydroxylase (TH), adrenergic receptor β1 (AR-β1) and CD68 was performed. Western blot was used to determine the protein expression level of TH, B type natriuretic peptide (BNP), and AR-β1. Results: Left ventricular ejection fraction (LVEF) declined continuously in HF group. LVEF was similar between denervated HF group and control group at various time points (P>0.05). LVEF was significantly higher in denervated HF group than in HF group at the end of week 4 (P<0.05). HE staining showed that cross sectional cardiomyocyte area was significantly larger in HF group than in control group and denervated HF group (P<0.05), which was similar between denervated HF group and control group (P>0.05). Masson staining showed that fibrosis level was significantly lower in denervated HF group than in HF group (P<0.05). IHC showed that TH+nerves and CD68⁺ macrophages were significantly increased in HF mice as compared to control mice (P<0.05), whereas this change was abolished in denervated HF group. AR-β1 was significantly down-regulated in HF group compared with control group (P<0.05), which was not affected by denervation (P>0.05). Western blot demonstrated that the expression level of TH and BNP was significantly higher in HF group compared with the control group (P<0.05), whereas this difference was diminished in denervated HF group (P>0.05). Conclusion: Bilateral superior cervical ganglionectomy can reduce sympathetic innervation and macrophage infiltration in pressure overloaded failure heart, thus attenuate cardiac remodeling and improve cardiac function.

Metabolic syndrome and renal cell cancer risk in Chinese males: a population-based prospective study

Background

Metabolic syndrome (MetS) is now a common public health problem. Few researches have reported the relationship between MetS and the risk of renal cell cancer (RCC). To investigate the association of metabolic syndrome and its components with the risk of RCC in Chinese males, the study was performed in the Kailuan male cohort, a large prospective cohort study.

Methods

A total of 104,333 eligible males enrolled in the every 2-year health checkup were involved in the Kailuan male cohort study (2006-2015). Information on demographic and socioeconomic characteristics, lifestyle, medical history and laboratory tests at baseline entry was obtained. Univariable and multivariable Cox proportional hazards regression models were used to estimate the association between MetS and the RCC risk.

Results

During a median follow-up of 8.9 years, 131 RCC cases were verified over a total of 824,211.96 person-years. Among the 5 single MetS components, hypertension (Systolic/diastolic blood pressure≥130/85 mm Hg or antihypertensive drug treatment of previously hypertension) (HR = 2.35, 95%CI:1.48-3.72) and elevated triglyceride (TG) (≥1.7mmol/L) (HR = 1.78, 95%CI:1.23-2.56) showed significant risk for RCC. Multivariate analysis showed that compared to those who did not meet MetS diagnostic criteria (number of abnormal MetS components&lt;3), HR of RCC risk for participants with MetS was 1.95 (95% CI 1.35-2.83). The number of abnormal MetS components was linearly associated with an increased risk of RCC (P trend&lt;0.001), and the HRs of RCC risk for males with 1, 2 and ≥3 MetS components were 1.27 (0.56-2.90), 2.42 (1.12-5.20) and 3.32 (1.56-7.07), respectively, compared with subjects without MetS components.

Conclusions

MetS was inversely associated with of RCC risk in males.

Key messages

MetS might be one of the scientific and important predictors of RCC. Controlling metabolic syndrome may potentially have key scientific and clinical significance for RCC prevention.

[Risk factors of retinopathy of prematurity in extremely low birth weight infants by strictly controlling oxygen inhalation after birth]

Objective: To observe the incidence and severity of retinopathy of prematurity (ROP) in extremely low birth weight (ELBW) infants by strictly controlling the risk factors of ROP, such as oxygen inhalation after birth, to explore the related factors of ROP in ELBW infants. Methods: This was a cross-sectional study. 166 ELBW infants underwent neonatal screening were enrolled in this study, whose birth weight was less than 1 000 g. There were 79 males and 87 females infants, whose average gestational age was (27.99±1.73)weeks, and average birth weight was (904.45±80.23)g. According to the final screening results, the ELBW infants were grouped as follows: (1)ROP group and non-ROP group; (2)severe ROP group and mild or no ROP group. Risk factors included gestational age, birth weight, test-tube infants, fetuses number, complications during pregnancy, delivery mode and Apgar scores in 1 to 10 minutes, weight and weight gain proportion at 1-6 weeks after birth, postnatal feeding mode, history of oxygen inhalation, anemia and blood transfusion, and other systemic diseases were recorded. And their correlation with severe ROP was analyzed by SPSS 20.0 statistical software. Results: Ninty-four (56.63%) ELBW infants developed ROP, 16 (9.64%) were severe ROP and 14(8.43%) received treatment. Average birth weight between ROP group (911.95±72.80)g and non-ROP group (894.67±88.58)g had no difference(t=1.379, P=0.170). Average gestational age between ROP group (27.49±1.53) weeks and non-ROP group (28.64±1.76) weeks had significant difference(t=-4.491,P<0.001).And pregnancy-induced hypertension during pregnancy (χ(2)=4.479, P=0.034), Apgar score in 5 minutes (t=-2.760, P=0.006) and 10 minutes (t=-2.099, P=0.043), pneumonia (χ(2)=6.233, P=0.013), neonatal pneumonia (χ(2)=18.026, P<0.001) had significant difference between ROP group and non-ROP group. There was no effect on weight (F=0.009,P=0.753) or weight gain proportion (F=2.394,P=0.124) at 1-6 weeks after birth in ELBW infants with or without ROP. Average birth weight between severe ROP group(875.63±74.85)g and mild or no ROP group(907.53±80.41)g had no difference(t=-1.518, P=0.131).Average gestational age between severe ROP group(26.88±1.31)weeks and mild or no ROP group (28.11±1.73)weeks had significant difference(t=-2.766,P=0.006).And only fundus hemorrhage (χ(2)=4.507,P=0.034) had significant difference between severe ROP group and mild or no ROP group. There was no effect on weight (F=2.683,P=0.103) or weight gain proportion (F=0.431,P=0.513) at 1-6 weeks after birth in ELBW infants with or without ROP. Logistic regression analysis revealed that only gestational age was correlated to the incidence (β=-0.437,P<0.001) and severity (β=-0.616,P=0.007) of ROP significantly. Conclusion: By strictly controlling the risk factors of ROP, such as oxygen inhalation after birth, the severe rate of ROP in ELBW infants is low. However, gestational age is still the inevitable independent high risk factor for the incidence of ROP in ELBW infants. (Chin J Ophthalmol, 2019, 55:280-288).

[Analysis of corneal and conjunctival sensitivities and its related factors of premature babies]

Objective: To analyse the corneal and conjunctival sensitivities of premature babies and to study the relevant influencing factors. Methods: Cross-sectional study. One hundred premature infants born at Women's Hospital School of Medicine Zhejiang University between May 2015 and September 2015 were enrolled, among which 51 were male (51%) and 49 were female (49%), the mean gestational age was (30.93±1.75)w, the mean corrected gestational age was (33.65±1.53)w, the mean birth weight was (1 592±336)g. The thresholds of cornea and conjunctiva of infants' left or right eyes were measured with Cochet-Bonnet aesthesiometer at 8-10 o'clock every morning when they naturally woke up, the minimum length of nylon wire that induced three successive times of eye-blink responses was recorded. Paired sample t test was used to compare the corneal and conjunctival sensitivities, the ocular surface sensitivities of preterm infants of different gender were compared using independent samples t-test, Pearson correlation and multiple linear regression analysis was conducted to analyze the correlation of corneal and conjuncitval sensitivities with gestational age, birth weight, age and corrected gestational age. Results: The mean corneal sensitivity was (44.85±5.53) mm and the mean conjunctival sensitivity was (23.50±5.48)mm in premature babies, corneal sensitivity was significantly higher than conjunctival sensitivity (t=25.620, P<0.001). No statistical significance was found between male and female preterm infants in corneal sensitivity [(44.80±5.83) mm vs. (44.90±5.25) mm, t=-0.085, P=0.933] and conjunctival sensitivity[(23.14±5.83) mm vs. (23.88±5.13) mm, t=-0.673, P=0.502]. Pearson correlation analysis showed that corneal sensitivity was significantly associated with conjunctival sensitivity in prematurity(r=0.676, P<0.001). There was significant correlation between corneal sensitivity and age, corrected gestational age (r=0.238, P=0.017; r=0.679, P<0.001), however no significant correlation was found between corneal sensitivity and gestational age, birth weight in preterm infants (r=0.067, P=0.510; r=-0.179, P=0.075). There was significant correlation between conjunctival sensitivity and corrected gestational age (r=0.490, P<0.001), however no significant correlation was found between conjunctival sensitivity and gestational age, birth weight and age in preterm infants (r=0.078, P=0.439; r=-0.096, P=0.344; r=0.151, P=0.133). Multiple linear regression revealed that corneal sensitivity(Y1) was positively correlated with corrected gestational age(X), the regression equation was Y1=2.45X-37.52, the conjunctical sensitivity(Y2) was also positively correlated with corrected gestational age(X), the regression equation was Y2=1.75X-35.41. Conclusions: The corneal sensitivity is higher than conjunctival sensitivity in premature babies.No statistical significance is found between male and female preterm infants in corneal sensitivity and conjunctival sensitivity. The corneal sensitivity and conjunctival sensitivity are correlated with corrected gestational age in preterm infants. The corneal and conjunctival sensitivities of premature babies tend to increase along with the increase of corrected gestational age. (Chin J Ophthalmol, 2018, 54: 115-119).

[Discussion on the related problems of pediatric burn treatment]

The incidence of pediatric burn is high currently. Many clinical problems in the treatment of pediatric burn are composed of fluid replacement during shock stage, wound treatment, nutrition and metabolism etc, which urgently need to be sorted out and updated again to make corresponding clinical guidelines, criteria, or consensus for standardizing the clinical diagnosis and treatment, so as to improve the clinical treatment level of pediatric burn.