MRI T_2 star mapping、T_1 images与3D DESS融合图在隐匿性膝关节软骨损伤中的应用

目的探讨T_2 star mapping、T_1 images与3D DESS融合伪彩图在关节软骨损伤中的诊断价值。方法对26例关节软骨损伤患者行T_2 star mapping、T_1 images和3D DESS扫描,并将T_1 images、T_2 star mapping与3D DESS图像融合,评价患者股骨、胫骨、髌骨关节软骨损伤程度并与关节镜结果对比,计算融合伪彩图诊断软骨损伤的特异性、敏感性及与关节镜诊断结果一致性。结果 T_1 images-3D DESS融合伪彩图诊断关节软骨损伤的敏感度、特异度及Kappa值分别为92.8%、93.0%、0.769,T_2 star mapping-3D DESS融合伪彩图诊断关节软骨损伤的敏感度、特异度及Kappa值分别为91.4%、94.2%、0.787。结论 T_2 star mapping、T_1 images与3D DESS融合伪彩图在关节软骨早期损伤评价上优于关节镜。

Efficacy of laparoscopic low anterior resection for colorectal cancer patients with 3D-vascular reconstruction for left coronary artery preservation

BACKGROUND Laparoscopic low anterior resection(LLAR)has become a mainstream surgical method for the treatment of colorectal cancer,which has shown many advantages in the aspects of surgical trauma and postoperative rehabilitation.However,the effect of surgery on patients'left coronary artery and its vascular reconstruction have not been deeply discussed.With the development of medical imaging technology,3D vascular reconstruction has become an effective means to evaluate the curative effect of surgery.AIM To investigate the clinical value of preoperative 3D vascular reconstruction in LLAR of rectal cancer with the left colic artery(LCA)preserved.METHODS A retrospective cohort study was performed to analyze the clinical data of 146 patients who underwent LLAR for rectal cancer with LCA preservation from January to December 2023 in our hospital.All patients underwent LLAR of rectal cancer with the LCA preserved,and the intraoperative and postoperative data were complete.The patients were divided into a reconstruction group(72 patients)and a nonreconstruction group(74 patients)according to whether 3D vascular reconstruction was performed before surgery.The clinical features,operation conditions,complications,pathological results and postoperative recovery of the two groups were collected and compared.RESULTS A total of 146 patients with rectal cancer were included in the study,including 72 patients in the reconstruction group and 74 patients in the nonreconstruction group.There were 47 males and 25 females in the reconstruction group,aged(59.75±6.2)years,with a body mass index(BMI)(24.1±2.2)kg/m^(2),and 51 males and 23 females in the nonreconstruction group,aged(58.77±6.1)years,with a BMI(23.6±2.7)kg/m^(2).There was no significant difference in the baseline data between the two groups(P>0.05).In the submesenteric artery reconstruction group,35 patients were type Ⅰ,25 patients were type Ⅱ,11 patients were type Ⅲ,and 1 patient was type Ⅳ.There were 37 type Ⅰ patients,24 type Ⅱ patients,12 type Ⅲ patients,and 1 type Ⅳ patient in the nonreconstruction group.There was no significant difference in arterial typing between the two groups(P>0.05).The operation time of the reconstruction group was 162.2±10.8 min,and that of the nonreconstruction group was 197.9±19.1 min.Compared with that of the reconstruction group,the operation time of the two groups was shorter,and the difference was statistically significant(t=13.840,P<0.05).The amount of intraoperative blood loss was 30.4±20.0 mL in the reconstruction group and 61.2±26.4 mL in the nonreconstruction group.The amount of blood loss in the reconstruction group was less than that in the control group,and the difference was statistically significant(t=-7.930,P<0.05).The rates of anastomotic leakage(1.4%vs 1.4%,P=0.984),anastomotic hemorrhage(2.8%vs 4.1%,P=0.672),and postoperative hospital stay(6.8±0.7 d vs 7.0±0.7 d,P=0.141)were not significantly different between the two groups.CONCLUSION Preoperative 3D vascular reconstruction technology can shorten the operation time and reduce the amount of intraoperative blood loss.Preoperative 3D vascular reconstruction is recommended to provide an intraoperative reference for laparoscopic low anterior resection with LCA preservation.

3D Shape Reconstruction of Lumbar Vertebra From Two X-ray Images and a CT Model

Structure reconstruction of 3 D anatomy from biplanar X-ray images is a challenging topic. Traditionally, the elastic-model-based method was used to reconstruct 3 D shapes by deforming the control points on the elastic mesh. However, the reconstructed shape is not smooth because the limited control points are only distributed on the edge of the elastic mesh.Alternatively, statistical-model-based methods, which include shape-model-based and intensity-model-based methods, are introduced due to their smooth reconstruction. However, both suffer from limitations. With the shape-model-based method, only the boundary profile is considered, leading to the loss of valid intensity information. For the intensity-based-method, the computation speed is slow because it needs to calculate the intensity distribution in each iteration. To address these issues, we propose a new reconstruction method using X-ray images and a specimen’s CT data. Specifically, the CT data provides both the shape mesh and the intensity model of the vertebra. Intensity model is used to generate the deformation field from X-ray images, while the shape model is used to generate the patient specific model by applying the calculated deformation field.Experiments on the public synthetic dataset and clinical dataset show that the average reconstruction errors are 1.1 mm and1.2 mm, separately. The average reconstruction time is 3 minutes.

A new projection model based robust 2D-3D registration method on Fourier-Mellin space for image guided intervention

An automatic method is proposed to solve the registration problem,which aligns a single 2D fluoroscopic image to a 3D image volume without demanding any additional media like calibration plate or user interactions.First,a mathematic projection model is designed which can reduce the influence of projection distortion on parameter optimization and improve the registration accuracy.Then,a two stage optimization method is proposed,which enables a robust registration in a wide parameter space.Furthermore,an automatic registration framework is proposed based on the FourierMellin robust image comparison descriptor.Experimental results show that the registration method has a high accuracy with average rotation error of 0.6 degree and average translation error of 1.4mm.

Clinical application of improved 2D computer-assisted fluoroscopic navigation through simulating a 3D vertebrae image to guide pedicle screw internal fixation

Objective To study the effect of using improved 2D computer-assisted fluoroscopic navigation through simulating 3D vertebrae image to guide pedicle screw internal fixation.Methods Posterior pedicle screw internal fixation,distraction

血清GSDMD、ACE2在小儿川崎病中的表达水平及临床意义研究

目的 探索血清焦孔素蛋白D(GSDMD)、血管紧张素转化酶2(ACE2)在小儿川崎病(KD)患儿中的表达水平及临床意义。方法 收集2020年1月至2022年1月该院诊治的90例KD患儿为KD组,根据KD患儿是否发生冠状动脉损伤(CAL),将KD组分为CAL组(32例)和非CAL组(58例),选取同期因急性呼吸道感染发热住该院的患儿50例为发热对照组,另选取同期该院儿外科行择期手术的腹股沟斜疝患儿50例为对照组。采用酶联免疫吸附试验检测血清GSDMD、ACE2水平。Pearson相关分析血清GSDMD、ACE2与临床指标的相关性。多因素Logisitic回归分析KD患者发生CAL的影响因素。受试者工作特征(ROC)曲线分析血清GSDMD、ACE2对KD患儿发生CAL的诊断价值。结果 KD组血清GSDMD、ACE2水平高于发热对照组和对照组,差异有统计学意义(均P<0.05)。相比于非CAL组,CAL组KD患儿发热持续时间、丙种球蛋白治疗时间、红细胞沉降率、血小板计数、C反应蛋白、GSDMD、ACE2水平均明显较高,而血钠、白蛋白明显较低,差异有统计学意义(均P<0.05)。Pearson相关分析结果,KD组患儿血清GSDMD、ACE2水平与发热持续时间、丙种球蛋白治疗时间、红细胞沉降率、血小板计数、C反应蛋白呈正相关(均P<0.05),与血钠、白蛋白呈负相关(均P<0.05)。多因素Logistic回归分析结果显示,血清GSDMD、ACE2升高是影响KD患儿发生CAL的独立危险因素。ROC曲线分析结果显示,血清GSDMD、ACE2两项联合检测KD患儿发生CAL的曲线下面积(AUC)及其95%CI为0.918(0.868~0.949),明显大于血清GSDMD、ACE2单项检测的AUC及其95%CI[依次为0.838(0.789~0.887)、0.865(0.811~0.912)],差异有统计学意义(Z=5.116、4.217,均P<0.05)。结论 血清GSDMD、ACE2两项联合检测对于KD患儿发生CAL具有较高的诊断价值。

ASME 2013 NDE中有关相控阵超声成像检测的要点评析 第二部分:计算机成像技术

探讨ASME最新版(2013)第Ⅴ卷《无损检测》第四章中有关承压设备相控阵超声成像检测的主要规定。评析承压设备焊接接头体积检测必须采用的CI(计算机成像)技术,突出相控阵超声检测扫查的典型模式,探测布图的典型示例,以及典型缺陷的相控阵读谱精要。意在对照国标国情,找差距、纠偏误,使中国企业正确执行ASME有关规范的水平更上一个台阶。

A Local Contrast Fusion Based 3D Otsu Algorithm for Multilevel Image Segmentation

To overcome the shortcomings of 1 D and 2 D Otsu’s thresholding techniques, the 3 D Otsu method has been developed.Among all Otsu’s methods, 3 D Otsu technique provides the best threshold values for the multi-level thresholding processes. In this paper, to improve the quality of segmented images, a simple and effective multilevel thresholding method is introduced. The proposed approach focuses on preserving edge detail by computing the 3 D Otsu along the fusion phenomena. The advantages of the presented scheme include higher quality outcomes, better preservation of tiny details and boundaries and reduced execution time with rising threshold levels. The fusion approach depends upon the differences between pixel intensity values within a small local space of an image;it aims to improve localized information after the thresholding process. The fusion of images based on local contrast can improve image segmentation performance by minimizing the loss of local contrast, loss of details and gray-level distributions. Results show that the proposed method yields more promising segmentation results when compared to conventional1 D Otsu, 2 D Otsu and 3 D Otsu methods, as evident from the objective and subjective evaluations.

Segmentation of retinal fluid based on deep learning:application of three-dimensional fully convolutional neural networks in optical coherence tomography images

AIM: To explore a segmentation algorithm based on deep learning to achieve accurate diagnosis and treatment of patients with retinal fluid.METHODS: A two-dimensional(2D) fully convolutional network for retinal segmentation was employed. In order to solve the category imbalance in retinal optical coherence tomography(OCT) images, the network parameters and loss function based on the 2D fully convolutional network were modified. For this network, the correlations of corresponding positions among adjacent images in space are ignored. Thus, we proposed a three-dimensional(3D) fully convolutional network for segmentation in the retinal OCT images.RESULTS: The algorithm was evaluated according to segmentation accuracy, Kappa coefficient, and F1 score. For the 3D fully convolutional network proposed in this paper, the overall segmentation accuracy rate is 99.56%, Kappa coefficient is 98.47%, and F1 score of retinal fluid is 95.50%. CONCLUSION: The OCT image segmentation algorithm based on deep learning is primarily founded on the 2D convolutional network. The 3D network architecture proposed in this paper reduces the influence of category imbalance, realizes end-to-end segmentation of volume images, and achieves optimal segmentation results. The segmentation maps are practically the same as the manual annotations of doctors, and can provide doctors with more accurate diagnostic data.

Three-dimensional arterial spin labeling and diffusion kurtosis imaging in evaluating perfusion and infarct area size in acute cerebral ischemia

BACKGROUND Early thrombolytic therapy is crucial to treat acute cerebral infarction,especially since the onset of thrombolytic therapy takes 1-6 h.Therefore,early diagnosis and evaluation of cerebral infarction is important.AIM To investigate the diagnostic value of magnetic resonance multi-delay threedimensional arterial spin labeling(3DASL)and diffusion kurtosis imaging(DKI)in evaluating the perfusion and infarct area size in patients with acute cerebral ischemia.METHODS Eighty-four patients who experienced acute cerebral ischemia from March 2019 to February 2021 were included.All patients in the acute stage underwent magnetic resonance-based examination,and the data were processed by the system’s own software.The apparent diffusion coefficient(ADC),average diffusion coefficient(MD),axial diffusion(AD),radial diffusion(RD),average kurtosis(MK),radial kurtosis(fairly RK),axial kurtosis(AK),and perfusion parameters post-labeling delays(PLD)in the focal area and its corresponding area were compared.The correlation between the lesion area of cerebral infarction under MK and MD and T2-weighted imaging(T2WI)was analyzed.RESULTS The DKI parameters of focal and control areas in the study subjects were compared.The ADC,MD,AD,and RD values in the lesion area were significantly lower than those in the control area.The MK,RK,and AK values in the lesion area were significantly higher than those in the control area.The MK/MD value in the infarct lesions was used to determine the matching situation.MK/MD<5 mm was considered matching and MK/MD≥5 mm was considered mismatching.PLD1.5s and PLD2.5s perfusion parameters in the central,peripheral,and control areas of the infarct lesions in MK/MD-matched and-unmatched patients were not significantly different.PLD1.5s and PLD2.5s perfusion parameter values in the central area of the infarct lesions in MK/MD-matched and-unmatched patients were significantly lower than those in peripheral and control areas.The MK and MD maps showed a lesion area of 20.08±5.74 cm^(2) and 22.09±5.58 cm^(2),respectively.T2WI showed a lesion area of 19.76±5.02 cm^(2).There were no significant differences in the cerebral infarction lesion areas measured using the three methods.MK,MD,and T2WI showed a good correlation.CONCLUSION DKI parameters showed significant difference between the focal and control areas in patients with acute ischemic cerebral infarction.3DASL can effectively determine the changes in perfusion levels in the lesion area.There was a high correlation between the area of the infarct lesions diagnosed by DKI and T2WI.

A study of the effect of Sudan I, III, and IV on the DNA/RNA ratio and 3D structure of HepG-2 using LCM

To explore the effect of sudan Ⅰ, Ⅲ, and Ⅳ on the DNA/RNA ratio and changes in the 3D structure of HepG-2. LCM and 3D images are used to detect the DNA/RNA ratio and changes in the 3D structure of HepG-2 when treated with different dosages of sudan Ⅰ, Ⅲ, and Ⅳ. The DNA/RNA ratio of the control group is 1. 223 2 ±0. 084 4, while the fluorescence intensity of DNA in HepG-2 treated with sudan Ⅰ, Ⅲ, and Ⅳ is markedly greater than that of RNA, with the low-dosage group showing significant effect （P 〈 0. 01 ）, yielding DNA/RNA ratios of 1. 609 6 ±0. 199 0, 1. 445 5 ±0. 163 3, 1. 708 1 ±0. 109 0 respectively; 3D images show that DNA fluorescence in HepG-2 is mostly concentrated in the nuclear region, and is denser and stronger than RNA fluorescence. The DNA/RNA ratio of a treated group increases after being treated with different dosages of sudan, but it declines with increasing dosage, and within a certain dosage range, sudan Ⅰ, Ⅲ, and Ⅳ are shown to promote the growth of HepG-2.

全脑3D-ASL和SWI联合检查在急性大面积脑梗死治疗中的价值

目的 探讨3D-ASL和SWI技术对急性大面积脑梗死治疗方案的指导价值。方法 回顾性分析46例急性大面积脑梗死患者在治疗前后的ASL和SWI表现,其中4例于治疗前、4例于动脉溶栓治疗后、5例静脉溶栓治疗后、33例抗凝治疗后、12例于治疗后30~60天复查,根据治疗方案的不同,分别统计急性大面积脑梗死患者ASL像高/低灌注、SWI像出血或侧枝循环形成的影像学表现,按NIHSS评分评价临床愈后效果。结果 本组46例中,14例患者ASL像上呈局部或大部分高灌注,占30%,其余32例为低灌注,占70%;4例行动脉溶栓,ASL像上呈局部或大部分高灌注,SWI像上均可见面积大小不等的低信号出血灶;5例行静脉溶栓,ASL像上呈高灌注3例、呈低灌注2例,SWI像可见脑出血1例;33例行抗凝治疗,7例于治疗后ASL像上呈高灌注,其中5例SWI像可见低信号出血灶,2例SWI像上未见出血灶;26例ASL像呈低灌注,10例SWI像上可见出血灶,16例SWI像未见出血灶。4例有房颤病史且平常口服抗凝药物的患者在抗凝治疗后均于ASL像上表现为高灌注。本组16例中,24例于SWI像上显示梗死区或周围血管增多、增粗,占52%。SWI像上9例可见沿大脑中动脉走行的条状低信号影,占20%。NIHSS评分：1例死亡,5例无变化,其余40例经治疗后均有不同程度好转。结论 联合运用3D-ASL和SWI技术指导急性大面积脑梗死患者的治疗,具有重要的临床价值。

长期补充维生素D对2型糖尿病患者冠状动脉钙化的影响研究

目的探讨2型糖尿病(T2DM)患者长期补充维生素D对冠状动脉钙化造成的影响。方法选取2017年3月至2020年3月本院收治的100例T2DM患者作为研究对象,采用随机数字法分为研究组与参照组,每组50例。参照组采用口服降糖药物和控制饮食治疗,研究组在参照组基础上联合补充维生素D治疗。比较两组治疗前后超敏C反应蛋白(hs-CRP)、血清25羟基维生素D_(3)[25-(OH)D_(3)]水平、血糖指标和冠状动脉钙化积分,并分析各相关指标与血清25-(OH)D_(3)的相关性。结果治疗后,两组HbA1c、FPG、BMI、HOMA-IR、hs-CRP均低于治疗前,且研究组均低于参照组,差异有统计学意义(P<0.05)。治疗后,两组冠状动脉积分均低于治疗前,且研究组低于参照组,差异有统计学意义(P<0.05)。治疗后,与血清25-(OH)D_(3)≥25 nmol/L的患者比较,血清25-(OH)D_(3)<25 nmol/L患者HbA1c、FPG、BMI、HOMA-IR、hs-CRP、冠状动脉钙化积分均明显较低(P<0.05)。血清25-(OH)D_(3)与HbA1c、FPG、BMI、HOMA-IR、hs-CRP、冠状动脉钙化积分均呈负相关(P<0.05)。结论T2DM患者长期补充维生素D有利于抑制炎性反应,对减轻冠状动脉钙化、降低胰岛素抵抗、控制血糖的稳定具有显著作用。

Auspicious Patterns and 3D Models Applied in the Creation of Personalized Products

Auspicious patterns are an important manifestation of traditional crafts aesthetics for Chinese culture, and it not only exhibits the clever tricks of folk art, showing more personality and characteristics of Chinese culture in the humanities and arts aesthetic concerns. It shows the traditional aesthetics, based on the harmonious and success, constructed by intelligence and humbleness, shaped by symmetry and balance. This thesis contains two topics: they are the 2D image materialization and the 3D model flattening. First is analyzing the image of the auspicious pattern, and transformed the 2D image into a solid model. The second is through the mathematical operation skills of the geometric model, the existing auspicious 3D model of the triangular mesh is scaled, appropriately rotated and divided to form a flattening model of different visual effects. Finally, these models by means of other modeling software were combined into a new 3D model, then through the 3D printer to quickly print out part of the unique personalized products, to promote the natural beauty of traditional Chinese culture.

Designing a High-Performance Deep Learning Theoretical Model for Biomedical Image Segmentation by Using Key Elements of the Latest U-Net-Based Architectures

Deep learning (DL) has experienced an exponential development in recent years, with major impact in many medical fields, especially in the field of medical image and, respectively, as a specific task, in the segmentation of the medical image. We aim to create a computer assisted diagnostic method, optimized by the use of deep learning (DL) and validated by a randomized controlled clinical trial, is a highly automated tool for diagnosing and staging precancerous and cervical cancer and thyroid cancers. We aim to design a high-performance deep learning model, combined from convolutional neural network (U-Net)-based architectures, for segmentation of the medical image that is independent of the type of organs/tissues, dimensions or type of image (2D/3D) and to validate the DL model in a randomized, controlled clinical trial. We used as a methodology primarily the analysis of U-Net-based architectures to identify the key elements that we considered important in the design and optimization of the combined DL model, from the U-Net-based architectures, imagined by us. Secondly, we will validate the performance of the DL model through a randomized controlled clinical trial. The DL model designed by us will be a highly automated tool for diagnosing and staging precancers and cervical cancer and thyroid cancers. The combined model we designed takes into account the key features of each of the architectures Overcomplete Convolutional Network Kite-Net (Kite-Net), Attention gate mechanism is an improvement added on convolutional network architecture for fast and precise segmentation of images (Attention U-Net), Harmony Densely Connected Network-Medical image Segmentation (HarDNet-MSEG). In this regard, we will create a comprehensive computer assisted diagnostic methodology validated by a randomized controlled clinical trial. The model will be a highly automated tool for diagnosing and staging precancers and cervical cancer and thyroid cancers. This would help drastically minimize the time and effort that specialists put into analyzing medical images, help to achieve a better therapeutic plan, and can provide a “second opinion” of computer assisted diagnosis.

Imaging Rock Density Distribution beneath Liwa Fracture Zone in the Southern Part of the Great Sumatran Fault System, Indonesia

We have imaged rock density distribution beneath Liwa fracture zone in the southern part of the the Sumatran Fault Zone by modelling and inverting Bouguer gravity data in two-and three-dimensional environments, respectively. The purpose of this study is aimed to figure out the subsurface distribution of rock densities associated with subsurface basement structure representing the evidence of trans-tensional tectonic product in the SF. In the gravity modeling, to eliminate distortions to the measured gravity values before modelling and inverting the data, Bouguer anomalies obtained in field measurements are reduced to the horizontal plane of z = +800 m as a representation of the average elevation in Liwa. For the inversion, we used algorithm implementing depth-and minimum volume weighting parameters in order to obtain a smooth model with better vertical resolution. The two-dimensional models show clearly surface topography of the basement rocks and the presence of normal faults. The reduced Bouguer anomaly of +800 m elevation shows the presence of structural lineaments extending primarily in a northwest-southeast direction, parallel to Sumatran Fault Zone and older graben faults showing a negative flower structure. From the three-dimensional inversion, the model illustrates an increase of density contrast, lower values being found near the surface and higher values in the deeper parts. The lower density contrast of 0.15 to 0.3 g/cm3 found in the rock groups at depths of 2 km and less is characteristic of relatively homogeneous and poorly compacted rocks. Rocks with moderate to high density contrast (>1.0 g/cm3) are recognized at depths of over 2 km. This model suggests a change of basement morphology as a function of depth, and delineates structural lineaments extending in northwest-southeast direction. This study supports the previous thought that Liwa area is underlain by graben structures, formed by trans-tensional tectonic events. Higher-density Tertiary volcanic breccia and lower-density Quaternary volcanic products of the Ranau Formation form the basement rocks and the overlying younger sediments, respectively.

An approach for converting orthographic projection drawings into 3D-CAD models

Abstract： Hand drawings and two dimensional （2D） CAD drawings have been replaced by three dimensional （3D） CAD models in mechanical design, but some 2D drawings produced before are needed in the new design. However, the techniques and software packages for automatically converting 2D drawings into 3D-CAD models with high precision have not yet been developed due to the difficulties to verify the validity of the drawings, to decide the hidden lines and eoncavo-convex faces, and to represent free-form surfaces. In addition, it is very time consuming to manually convert a large number of 2D drawings into 3D CAD models. To address these problems, we propose an approach for converting 2D drawings into 3D-CAD models automatically.

3D laparoscopic anatomical hepatectomy guided by 2D real-time indocyanine green fluorescence imaging for hepatocellular carcinoma

Laparoscopic anatomical hepatectomy(LAH)for patients with hepatocellular carcinoma(HCC)has been advocated by many surgeons in the hope of producing better oncological outcomes.Two recent techniques,3D laparoscopic system and 2D real-time indocyanine green fluorescence imaging(r-ICG)guidance,are benefit for improving the operative precision of LAH in different aspects.However,these two techniques cannot be applied concomitantly because of the technical limitation.Although a new modern laparoscopic system with both 3D and indocyanine green(ICG)imaging mode has been designed,it has not been listed in many countries including China.Thus,we design a new procedure to perform the 3D LAH with 2D r-ICG guidance for HCCs with conventional laparoscopic systems.In this procedure,both 3D and 2D laparoscopic systems were used.A total of 11 patients with HCC received 3D laparoscopic right posterior sectionectomy(LRPS)with 2D r-ICG guidance.The right posterior Glissonian pedicle was clamped under the 3D vision.Then ICG solution was then intravenously administrated.The liver parenchyma was transected under the 3D vision and guided by 2D ICG vision simultaneously.There was no severe complications(Clavien-Dindo≥III)and operation related death.The 90-day mortality was also nil.By using this procedure,the advantages of two techniques,3D laparoscopic system and 2D r-ICG guidance,were combined so that LAH could be performed with more precision.However,it should be validated in more studies.

A Systematic Literature Review of Deep Learning Algorithms for Segmentation of the COVID-19 Infection

Coronavirus has infected more than 753 million people,ranging in severity from one person to another,where more than six million infected people died worldwide.Computer-aided diagnostic(CAD)with artificial intelligence(AI)showed outstanding performance in effectively diagnosing this virus in real-time.Computed tomography is a complementary diagnostic tool to clarify the damage of COVID-19 in the lungs even before symptoms appear in patients.This paper conducts a systematic literature review of deep learning methods for classifying the segmentation of COVID-19 infection in the lungs.We used the methodology of systematic reviews and meta-analyses(PRISMA)flow method.This research aims to systematically analyze the supervised deep learning methods,open resource datasets,data augmentation methods,and loss functions used for various segment shapes of COVID-19 infection from computerized tomography(CT)chest images.We have selected 56 primary studies relevant to the topic of the paper.We have compared different aspects of the algorithms used to segment infected areas in the CT images.Limitations to deep learning in the segmentation of infected areas still need to be developed to predict smaller regions of infection at the beginning of their appearance.

Radial Hahn Moment Invariants for 2D and 3D Image Recognition

Recently, orthogonal moments have become efficient tools for two-dimensional and three-dimensional（2D and 3D） image not only in pattern recognition, image vision, but also in image processing and applications engineering. Yet, there is still a major difficulty in 3D rotation invariants. In this paper, we propose new sets of invariants for 2D and 3D rotation, scaling and translation based on orthogonal radial Hahn moments. We also present theoretical mathematics to derive them. Thus, this paper introduces in the first case new 2D radial Hahn moments based on polar representation of an object by one-dimensional orthogonal discrete Hahn polynomials, and a circular function. In the second case, we present new 3D radial Hahn moments using a spherical representation of volumetric image by one-dimensional orthogonal discrete Hahn polynomials and a spherical function. Further 2D and 3D invariants are derived from the proposed 2D and 3D radial Hahn moments respectively, which appear as the third case. In order to test the proposed approach, we have resolved three issues： the image reconstruction, the invariance of rotation, scaling and translation, and the pattern recognition. The result of experiments show that the Hahn moments have done better than the Krawtchouk moments, with and without noise. Simultaneously, the mentioned reconstruction converges quickly to the original image using 2D and 3D radial Hahn moments, and the test images are clearly recognized from a set of images that are available in COIL-20 database for 2D image, and Princeton shape benchmark（PSB） database for 3D image.