Quality of Images Reconstructed by Deep Learning Reconstruction Algorithm for Head and Neck CT Angiography at 100 kVp / 中国医学科学院学报

Objective To evaluate the impact of deep learning reconstruction algorithm on the image quality of head and neck CT angiography (CTA) at 100 kVp. Methods CT scanning was performed at 100 kVp for the 37 patients who underwent head and neck CTA in PUMC Hospital from March to April in 2021.Four sets of images were reconstructed by three-dimensional adaptive iterative dose reduction (AIDR 3D) and advanced intelligent Clear-IQ engine (AiCE) (low,medium,and high intensity algorithms),respectively.The average CT value,standard deviation (SD),signal-to-noise ratio (SNR),and contrast-to-noise ratio (CNR) of the region of interest in the transverse section image were calculated.Furthermore,the four sets of sagittal maximum intensity projection images of the anterior cerebral artery were scored (1 point:poor,5 points:excellent). Results The SNR and CNR showed differences in the images reconstructed by AiCE (low,medium,and high intensity) and AIDR 3D (all P<0.01).The quality scores of the image reconstructed by AiCE (low,medium,and high intensity) and AIDR 3D were 4.78±0.41,4.92±0.27,4.97±0.16,and 3.92±0.27,respectively,which showed statistically significant differences (all P<0.001). Conclusion AiCE outperformed AIDR 3D in reconstructing the images of head and neck CTA at 100 kVp,being capable of improving image quality and applicable in clinical examinations.

Optimal Parameters for Virtual Mono-Energetic Imaging of Liver Solid Lesions / 中国医学科学院学报

Objective To explore the optimal parameters for virtual mono-energetic imaging of liver solid lesions. Methods A retrospective analysis was performed on 60 patients undergoing contrast-enhanced spectral CT of the abdomen.The iodine concentration values of hepatic arterial phase images and the CT values of different mono-energetic images were measured.The correlation coefficient and coefficient of variation were calculated. Results The average correlation coefficients between iodine concentrations and CT values of hepatic solid lesion images at 40,45,50,55,60,65,and 70 keV were 0.996,0.995,0.993,0.989,0.978,0.970,and 0.961,respectively.The correlation coefficients at 40(P=0.007),45(P=0.022),50 keV (P=0.035)were higher than that at 55 keV,and the correlation coefficients at 40 keV(P=0.134) and 45 keV(P=0.368) had no significant differences from that at 50 keV.The coefficients of variation of the CT values at 40,45,and 50 keV were 0.146,0.154,and 0.163,respectively. Conclusion The energy of 40 keV is optimal for virtual mono-energetic imaging of liver solid lesions in the late arterial phase,which is helpful for the diagnosis of liver diseases.

The performance of digital chest radiographs in the detection and diagnosis of pulmonary nodules and the consistency among readers / 中华肿瘤杂志

Objective: To investigate the detection and diagnostic efficacy of chest radiographs for ≤30 mm pulmonary nodules and the factors affecting them, and to compare the level of consistency among readers. Methods: A total of 43 patients with asymptomatic pulmonary nodules who consulted in Cancer Hospital, Chinese Academy of Medical Sciences from 2012 to 2014 and had chest CT and X-ray chest radiographs during the same period were retrospectively selected, and one nodule ≤30 mm was visible on chest CT images in the whole group (total 43 nodules in the whole group). One senior radiologist with more than 20 years of experience in imaging diagnosis reviewed CT images and recording the size, morphology, location, and density of nodules was selected retrospectively. Six radiologists with different levels of experience (2 residents, 2 attending physicians and 2 associate chief physicians independently reviewed the chest images and recorded the time of review, nodule detection, and diagnostic opinion. The CT imaging characteristics of detected and undetected nodules on X images were compared, and the factors affecting the detection of nodules on X-ray images were analyzed. Detection sensitivity and diagnosis accuracy rate of 6 radiologists were calculated, and the level of consistency among them was compared to analyze the influence of radiologists' seniority and reading time on the diagnosis results. Results: The number of nodules detected by all 6 radiologists was 17, with a sensitivity of detection of 39.5%(17/43). The number of nodules detected by ≥5, ≥4, ≥3, ≥2, and ≥1 physicians was 20, 21, 23, 25, and 28 nodules, respectively, with detection sensitivities of 46.5%, 48.8%, 53.5%, 58.1%, and 65.1%, respectively. Reasons for false-negative result of detection on X-ray images included the size, location, density, and morphology of the nodule. The sensitivity of detecting ≤30 mm, ≤20 mm, ≤15 mm, and ≤10 mm nodules was 46.5%-58.1%, 45.9%-54.1%, 36.0%-44.0%, and 36.4% for the 6 radiologists, respectively; the diagnosis accuracy rate was 19.0%-85.0%, 16.7%-6.5%, 18.2%-80.0%, and 0%-75.0%, respectively. The consistency of nodule detection among 6 doctors was good (Kappa value: 0.629-0.907) and the consistency of diagnostic results among them was moderate or poor (Kappa value: 0.350-0.653). The higher the radiologist's seniority, the shorter the time required to read the images. The reading time and the seniority of the radiologists had no significant influence on the detection and diagnosis results (P>0.05). Conclusions: The ability of radiographs to detect lung nodules ≤30 mm is limited, and the ability to determine the nature of the nodules is not sufficient, and the increase in reading time and seniority of the radiologists will not improve the diagnostic accuracy. X-ray film exam alone is not suitable for lung cancer diagnosis.

A three dimensional convolutional neural network pulmonary nodule detection algorithm based on the multi-scale attention mechanism / 生物医学工程学杂志

Early screening based on computed tomography (CT) pulmonary nodule detection is an important means to reduce lung cancer mortality, and in recent years three dimensional convolutional neural network (3D CNN) has achieved success and continuous development in the field of lung nodule detection. We proposed a pulmonary nodule detection algorithm by using 3D CNN based on a multi-scale attention mechanism. Aiming at the characteristics of different sizes and shapes of lung nodules, we designed a multi-scale feature extraction module to extract the corresponding features of different scales. Through the attention module, the correlation information between the features was mined from both spatial and channel perspectives to strengthen the features. The extracted features entered into a pyramid-similar fusion mechanism, so that the features would contain both deep semantic information and shallow location information, which is more conducive to target positioning and bounding box regression. On representative LUNA16 datasets, compared with other advanced methods, this method significantly improved the detection sensitivity, which can provide theoretical reference for clinical medicine.

Cone-beam computed tomography cinematic rendering: clinical, teaching and research applications

Abstract Cone-beam computed tomography (CBCT) is an essential imaging method that increases the accuracy of diagnoses, planning and follow-up of endodontic complex cases. Image postprocessing and subsequent visualization relies on software for three-dimensional navigation, and application of indexation tools to provide clinically useful information according to a set of volumetric data. Image postprocessing has a crucial impact on diagnostic quality and various techniques have been employed on computed tomography (CT) and magnetic resonance imaging (MRI) data sets. These include: multiplanar reformations (MPR), maximum intensity projection (MIP) and volume rendering (VR). A recent advance in 3D data visualization is the new cinematic rendering reconstruction method, a technique that generates photorealistic 3D images from conventional CT and MRI data. This review discusses the importance of CBCT cinematic rendering for clinical decision-making, teaching, and research in Endodontics, and a presents series of cases that illustrate the diagnostic value of 3D cinematic rendering in clinical care.

Reliability of Linear and Curvilinear Measurements on Cone-Beam Computed Tomography Images for the Evaluation of Implant Sites and Jaw Pathologies

ABSTRACT Objective: To evaluate the intra-examiner and inter-examiner reliability of linear and curvilinear measurements for the complete assessment of implant sites and jaw pathologies using Cone-Beam Computed Tomography (CBCT). Material and Methods: Fifty cone-beam computed tomographic images of patients were retrieved from the archives of Dentomaxillofacial Radiology. CBCT images taken for implant planning and evaluation of intrabony jaw pathologies (benign cyst/tumor) were included. Two expert oral and maxillofacial radiologists analyzed the images independently and made the measurements. The images for implant planning were analyzed for width, the height of the edentulous site, and the qualitative analysis of bone in the region. Jaw pathologies were assessed for linear dimensions and curvilinear measurements. Results: The inter-observer measurement error for implant site analysis ranged from 0.12 to 0.42 mm with almost perfect agreement (ICC: 0.94 to 1). The inter-observer measurement error for jaw pathology was 0.09 to 0.25 mm (ICC: 0.98-1). Curvilinear measurements showed perfect agreement between the observers. The intraobserver reliability for the various parameters used for the assessment of the implant site and jaw pathologies indicated almost perfect agreement. Conclusion: Reliability between the radiologists is high for various measurements on CBCT images taken for implant planning and jaw pathologies.

Averaging Strategy to Form the Imaging for Routine Reading of Insulinoma from Pancreatic Perfusion Dataset / 中国医学科学院学报

Objective To determine the appropriate averaging strategy for pancreatic perfusion datasets to create images for routine reading of insulinoma.Methods Thirty-nine patients undergoing pancreatic perfusion CT in Peking Union Medical College Hospital and diagnosed as insulinoma by pathology were enrolled in this retrospective study.The time-density curve of abdominal aorta calculated by software dynamic angio was used to decide the timings for averaging.Five strategies,by averaging 3,5,7,9 and 11 dynamic scans in perfusion,all including peak enhancement of the abdominal aorta,were investigated in the study.The image noise,pancreas signal-to-noise ratio(SNR),lesion contrast and lesion contrast-to-noise ratio(CNR)were recorded and compared.Besides,overall image quality and insulinoma depiction were also compared.ANOVA and Friedman's test were performed.Results The image noise decreased and the SNR of pancreas increased with the increase in averaging time points(all P0.99)and were higher than that of the first group(all P<0.05).There was no significant difference in overall image quality among the 5 groups(P=0.977).Conclusions Image averaged from 5 scans showed moderate image noise,pancreas SNR and relatively high lesion contrast and lesion CNR.Therefore,it is advised to be used in image averaging to detect insulinoma.

Técnicas de segmentación y procesamiento para la detección de Carcinomas Renales en imágenes de Tomografía Abdominal / Segmentation and processing techniques for the detection of Renal Carcinomas in Abdominal Tomography images

Una de las campañas más reconocidas en el mundo es la lucha contra el cáncer, siendo el sistema renal uno de los más afectados por esta patología. El carcinoma de células renales (CCR), el más común de cáncer renal en los adultos, representa la sexta causa de muerte por cáncer. Debido al aumento en el uso de las técnicas de diagnóstico por imagen, las lesiones renales pueden ser diagnosticadas en forma incidental aproximadamente en 50% de los casos. Cuba apuesta por el uso de la tecnología en la salud y en la Universidad de las Ciencias Informáticas (UCI) se ha desarrollado un sistema para el almacenamiento, transmisión y visualización de imágenes médicas (XAVIA PACS), el cual se encuentra implantado en varios hospitales del país, pero no cuenta con alternativas para realizar la detección del CCR en imágenes tomográficas, haciendo más lento el diagnóstico, lo que se traduce en menos posibilidades para el paciente. La presente investigación tiene como objetivo realizar un análisis sobre las principales técnicas de segmentación y procesamiento para la detección de carcinomas renales en imágenes de tomografías abdominal, que propicie a los equipos de desarrollo contar con la base teórica necesaria para enfrentar el problema en cuestión. Para ello se realizó un análisis documental sobre trabajos relacionados con la temática y que propician soluciones al problema. Se estudiaron algoritmos y técnicas computacionales efectivas para la segmentación y procesamiento de imágenes abdominales. Como resultado de la investigación se obtuvieron los algoritmos más acordes para el sistema XAVIA PACS y el contexto médico cubano(AU)

One of the most recognized campaigns in the world is the fight against cancer, the kidney system being one of the most affected by this pathology. Renal cell carcinoma (RCC), the most common form of kidney cancer in adults, represents the sixth leading cause of cancer death. Due to the increased use of diagnostic imaging techniques, kidney injuries can be diagnosed incidentally in approximately 50% of cases. Cuba is committed to the use of technology in health and a system for the storage, transmission and display of medical images (XAVIA PACS) has been developed at the University of Computer Sciences (UCI), which is implanted in several hospitals of the country, but it does not have alternatives to detect RCC in tomographic images, slowing down the diagnosis, which translates into fewer possibilities for the patient. The objective of this research is to carry out an analysis on the main segmentation and processing techniques for the detection of renal carcinomas in abdominal tomography images, which provides development teams with the theoretical basis necessary to face the problem in question. For this, a documentary analysis was carried out on works related to the subject and that provide solutions to the problem. Algorithms and effective computational techniques for the segmentation and processing of abdominal images were studied. As a result of the research, the most suitable algorithms for the XAVIA PACS system and the Cuban medical context were obtained(AU)

Soft Tissue Cephalometric Measurements Among Malaysian Malays and Chinese

Abstract Objective: To obtain the standardized values of individuals of Malaysian Malay and Chinese for further relevant research, such as treatment planning and aesthetical considerations. Material and Methods: In this retrospective study, 440 (305 were Malays and 135 were Chinese) standardized lateral cephalometric radiographs of orthodontic patients selected through simple random sampling are profiled using Holdaway's analysis. The independent t-test was used to assess the disparities in race and gender. The significant level was p<0.05. Results: Significant differences were found between the Malays and Chinese in their skeletal profile convexity, superior sulcus depth, inferior sulcus to the H line and nose prominence. Between Malay females and males, there are significant differences in superior sulcus depth, soft tissue subnasale to H line, basic upper lip thickness, upper lip thickness and nose prominence. Between Chinese males and females, there were differences in their skeletal profile convexity, upper lip to H line, basic upper lip thickness and upper lip thickness. Conclusion: The findings demonstrated the difference between standardized norms and the unique profiles of Malaysian Malays and Chinese. There are significant gender disparities in the soft tissue cephalometric measurements among Malaysian Malay and Chinese subjects.

Optimal Mono-energy in Mono-energetic Spectral Computed Tomography of Enhanced Renal Cortex in Cortical Phase Based on Iodine Concentration / 中国医学科学院学报

Objective To identify the optimal mono-energetic enhanced spectral CT for renal cortex in cortical phase based on the iodine concentration. Methods Fifty patients with normal renal function received the abdominal enhanced spectral CT examination.The iodine concentration and CT values of the multiple mono-energetic spectral images were measured on renal cortex in cortical phase,and the correlation between the iodine concentration and the CT values and the coefficient of variation(CV)were analyzed. Results The correlation analysis demonstrated that the correlation coefficient was 0.994,0.994,0.993,0.987,0.976,0.960,and 0.938 between mono-energetic spectral CT images(40-100 keV with interval 10 keV,respectively)and iodine concentration(all

Effect of Monochromatic Energy Image Synthesized from Dual-layer Detector Spectral CT on Imaging of Inferior Vena Cava / 中国医学科学院学报

To evaluate the effect of monochromatic energy image on inferior vena cava imaging quality on dual-layer detector spectral CT. Totally 39 patients who were clinically suspected of abdominal disease and referred to perform contrast-enhanced computed tomography(CT)were prospectively enrolled and underwent abdominal examination using a single-source,dual-detector spectral CT.The delayed phase scan was performed 3 minutes after injection of 60 ml of iopamidol(320 mg/ml)at a rate of 3 ml/s.The raw images were reconstructed to obtain conventional mixed energy images and spectral based images(SBI).The 40,50,60,and 70 keV single energy images were obtained.The CT value,noise,and signal-to-noise(SNR)of inferior vena cava and the contrast-to-noise(CNR)of inferior vena cava relative to psoas major on conventional mixed energy images and the 40,50,60,70 keV single energy images were measured.The SNRs and CNRs on monoenergetic 40-70 keV images were compared with polychromatic 120 kVp images.ANOVA was used to compare the CT value,noise,SNR,and CNR among these five groups.The optimal monoenergetic image set was chosen. The differences in CT value,noise,SNR,CNR of inferior vena cava were statistically significant among five groups(all <0.05).The SNR and CNR in 40 keV group and 50 keV group were significantly higher than those in other groups(all <0.05).The SNR of 40 keV group was significantly higher than that of 50 keV group(=0.002).The CNR of 40 keV group was not statistical different compared with that of 50 keV group(=0.630). 40 keV is the optimal monoenergetic energy level for the inferior vena cava on dual-layer detector spectral CT and may be valuable for the diagnosis of inferior vena cava disease.

Patency of individual and sequential coronary artery bypass in patients with ischemic heart disease: a meta-analysis

Abstract Objective: To evaluate the patency of individual and sequential coronary artery bypass in patients with ischemic heart disease. Methods: We searched PubMed, Cochrane Library, Excerpta Medica Database, and ClinicalTrials.gov databases for controlled trials. Endpoints included graft patency, anastomosis patency, occluded rates in left anterior descending (LAD) system and right coronary artery (RCA) system, in-hospital mortality, and follow-up mortality. Pooled risk ratios (RRs) and standardized mean difference (SMD) were used to assess the relative data. Results: Nine cohorts, including 7100 patients and 1440 grafts under individual or sequential coronary artery bypass. There were no significant differences between individual and sequential coronary artery bypass in the graft patency (RR=0.96; 95% CI=0.91-1.02; P=0.16; I2=87%), anastomosis patency (RR=0.95; 95% CI=0.91-1.00; P=0.05; I2=70%), occluded rate in LAD system (RR=1.03; 95% CI=0.92-1.16; P=0.58; I2=37%), occluded rate in RCA system (RR=1.36; 95% CI=0.72-2.57; P=0.35; I2=95%), in-hospital mortality (RR=1.57; 95% CI=0.92-2.69; P=0.10; I2=0%), and follow-up mortality (RR=0.96; 95% CI=0.36-2.53; P=0.93; I2=0%). Conclusion: No significant differences on clinical data were observed regarding anastomosis patency, occluded rate in LAD system, occluded rate in RCA system, in-hospital mortality, and follow-up mortality, indicating that the patency of individual and the patency of sequential coronary artery bypass are similar to each other.

Researching on the Classification of Computer Aided Diagnosis/Detection Software / 中国医疗器械杂志

Based on the developing situation of Computer Aided Diagnosis/Detection (CAD) software, considering the domestic and international regulation of CAD software, according to current Medical Device Classification Catalog and related laws of China Food and Drug Administration (CFDA), this paper investigated and analyzed the classification of CAD software, and provided technical suggestion on classifying principle of CAD software applying Artificial Intelligence (AI) or other advanced technology from medical device regulation scope, for the reference of regulatory and technical departments.

Impact of optional reconstruction kernel on image quality of pulmonary ground glass nodules using the third generation dual-source computed tomography / 中南大学学报(医学版)

To explore the value of the third generation dual-source computed tomography (CT) convolution kernel in display of pulmonary ground-glass nodule (GGN) in transverse image reconstruction.
 Methods: A total of 52 lung adenocarcinoma patients with lung CT data were selected from February 2018 to January 2019 for this study. The pulmonary CT data were reconstructed by convolutional nucleus B157, Br54, and Br49. The signal-to-noise ratio (SNR), the contrast-to-noise ratio (CNR), and the standard deviation (SD) of the image at the GGN were used as the objective evaluation standard of image quality. Subjective image quality was scored by 2 radiologists from 3 aspects (overall image quality, noise, and lesion outline).
 Results: Objective image quality evaluation, SNR and CNR of reconstructed convolution kernel Br49 (SNR: 11.36±5.39, CNR: 7.19±4.29), Br54 (SNR: 8.30±3.35, CNR: 5.09±2.86) are greater than those of Bl57 (SNR: 4.18±2.10, CNR: 3.25±1.78; all P<0.01). SD of reconstructed convolution kernel Br49 (61.80±20.17) and Br54 (80.45±20.31) is smaller than that of Bl57 (137.92±31.11, both P<0.01). In the subjective image quality evaluation, the overall image quality score 5.0(4.5, 5.0) of Br54 was higher than that of all other images [Br49: 3.0(3.0, 4.0), Bl57: 3.0(3.0, 3.5); both P<0.05]. The Br54 image showed that the lesion contour ability score 5.0(4.0, 5.0) was higher than all other images [Br49: 4.0(4.0, 5.0), Bl57: 3.0(3.0, 3.0); both P<0.05]; Br49 image noise score 3.0(3.0, 3.0) is the lowest one [Br54 4.0(4.0, 4.0), Bl57 5.0(5.0, 5.0); both P<0.05].
 Conclusion: The reasonable selection of CT convolution kernel plays an important role in the subjective and objective image quality of GGN. It is suggested that Br54 should be used as the reconstruction of convolutional kernel in pulmonary ground glass nodules, which is helpful for doctors to find and diagnose GGN.

Pulmonary nodule detection method based on convolutional neural network / 生物医学工程学杂志

A method was proposed to detect pulmonary nodules in low-dose computed tomography (CT) images by two-dimensional convolutional neural network under the condition of fine image preprocessing. Firstly, CT image preprocessing was carried out by image clipping, normalization and other algorithms. Then the positive samples were expanded to balance the number of positive and negative samples in convolutional neural network. Finally, the model with the best performance was obtained by training two-dimensional convolutional neural network and constantly optimizing network parameters. The model was evaluated in Lung Nodule Analysis 2016(LUNA16) dataset by means of five-fold cross validation, and each group's average model experiment results were obtained with the final accuracy of 92.3%, sensitivity of 92.1% and specificity of 92.6%.Compared with other existing automatic detection and classification methods for pulmonary nodules, all indexes were improved. Subsequently, the model perturbation experiment was carried out on this basis. The experimental results showed that the model is stable and has certain anti-interference ability, which could effectively identify pulmonary nodules and provide auxiliary diagnostic advice for early screening of lung cancer.

Lung nodule segmentation based on fuzzy c-means clustering and improved random walk algorithm / 生物医学工程学杂志

Accurate segmentation of pulmonary nodules is an important basis for doctors to determine lung cancer. Aiming at the problem of incorrect segmentation of pulmonary nodules, especially the problem that it is difficult to separate adhesive pulmonary nodules connected with chest wall or blood vessels, an improved random walk method is proposed to segment difficult pulmonary nodules accurately in this paper. The innovation of this paper is to introduce geodesic distance to redefine the weights in random walk combining the coordinates of the nodes and seed points in the image with the space distance. The improved algorithm is used to achieve the accurate segmentation of pulmonary nodules. The computed tomography (CT) images of 17 patients with different types of pulmonary nodules were selected for segmentation experiments. The experimental results are compared with the traditional random walk method and those of several literatures. Experiments show that the proposed method has good accuracy in the segmentation of pulmonary nodule, and the accuracy can reach more than 88% with segmentation time is less than 4 seconds. The results could be used to assist doctors in the diagnosis of benign and malignant pulmonary nodules and improve clinical efficiency.

Risk assessment of hepatocellular carcinoma development for indeterminate hepatic nodules in patients with chronic hepatitis B

BACKGROUND/AIMS: A risk prediction model for the development of hepatocellular carcinoma (HCC) from indeterminate nodules detected on computed tomography (CT) (Rad(CT) score) in patients with chronic hepatitis B (CHB)-related cirrhosis was proposed. We validated this model for indeterminate nodules on magnetic resonance imaging (MRI).METHODS: Between 2013 and 2016, Liver Imaging Reporting and Data System (LI-RADS) 2/3 nodules on MRI were detected in 99 patients with CHB. The Rad(CT) score was calculated.RESULTS: The median age of the 72 male and 27 female subjects was 58 years. HCC history and liver cirrhosis were found in 47 (47.5%) and 44 (44.4%) patients, respectively. The median Rad(CT) score was 112. The patients with HCC (n=41, 41.4%) showed significantly higher Rad(CT) scores than those without (median, 119 vs. 107; P=0.013); the Chinese university-HCC and risk estimation for HCC in CHB (REACH-B) scores were similar (both P>0.05). Arterial enhancement, T2 hyperintensity, and diffusion restriction on MRI were not significantly different in the univariate analysis (all P>0.05); only the Rad(CT) score significantly predicted HCC (hazard ratio [HR]=1.018; P=0.007). Multivariate analysis showed HCC history was the only independent HCC predictor (HR=2.374; P=0.012). When the subjects were stratified into three risk groups based on the Rad(CT) score (<60, 60–105, and >105), the cumulative HCC incidence was not significantly different among them (all P>0.05, log-rank test).CONCLUSIONS: HCC history, but not Rad(CT) score, predicted CHB-related HCC development from LI-RADS 2/3 nodules. New risk models optimized for MRI-defined indeterminate nodules are required.

Research progress on computed tomography image detection and classification of pulmonary nodule based on deep learning / 生物医学工程学杂志

Computer-aided diagnosis based on computed tomography (CT) image can realize the detection and classification of pulmonary nodules, and improve the survival rate of early lung cancer, which has important clinical significance. In recent years, with the rapid development of medical big data and artificial intelligence technology, the auxiliary diagnosis of lung cancer based on deep learning has gradually become one of the most active research directions in this field. In order to promote the deep learning in the detection and classification of pulmonary nodules, we reviewed the research progress in this field based on the relevant literatures published at domestic and overseas in recent years. This paper begins with a brief introduction of two widely used lung CT image databases: lung image database consortium and image database resource initiative (LIDC-IDRI) and Data Science Bowl 2017. Then, the detection and classification of pulmonary nodules based on different network structures are introduced in detail. Finally, some problems of deep learning in lung CT image nodule detection and classification are discussed and conclusions are given. The development prospect is also forecasted, which provides reference for future application research in this field.

Development of a new cone-beam computed tomography software for endodontic diagnosis

Abstract Cone-beam computed tomography (CBCT) has promoted changes in approaches in Endodontics, and enhanced decision-making in complex clinical cases. Despite the technological advancements in CBCT hardware, the interpretation of the acquired images is still compromised by viewing software packages that often have limited navigational tools and lack adequate filters to overcome some challenges of the CBCT technology such as artefacts. This study reviews the current limitations of CBCT and the potential of a new CBCT software package (e-Vol DX, CDT- Brazil) to overcome these aspects and support diagnosing, planning and managing of endodontic cases. This imaging method provide high resolution images due to submillimeter voxel sizes, dynamic multi-plane imaging navigation and ability to change the volume parameters such as slice thickness and slice intervals and data correction applying imaging filters and manipulating brightness and contrast. The main differences between e-Vol DX and other software packages are: compatibility with all current CBCT scanners with the capacity to export DICOM Data, a more comprehensive brightness and contrast library, as other applications, in which adjustments are limited, do not usually support all the DICOM dynamic range features; Custom slice thickness adjustment, often limited and pre-defined in other applications; Custom Sharpening adjustment, often limited in other applications; advanced noise reduction algorithm that enhances image quality; preset imaging filters, dedicated endodontic volume rendering filters with the ability to zoom the image over 1000x (3D reconstructions) without loss of resolution and automatic imaging parameters customization for better standardization and opportunities for research; capture screen resolution of 192 dpi, with a 384 dpi option, in contrast to the 96 dpi of most similar applications. This new CBCT software package may support decision-making for the treatment of complex endodontic cases and improve diagnosis and treatment results. Effective improvement of image quality favors the rational prescription and interpretation of CBCT scans.

Resumo A tomografia computadorizada de feixe cônico (TCFC) promoveu mudanças nas abordagens de Endodontia e melhorou a tomada de decisões em casos clínicos complexos. Apesar dos avanços tecnológicos no hardware da TCFC, a interpretação da imagem adquirida ainda é comprometida pela visualização dos softwares, que muitas vezes têm ferramentas de navegação limitadas e falta de filtros adequados para superar estes desafios, como artefatos. Este estudo analisa as limitações atuais da TCFC e o potencial de um novo software (e-Vol DX, CDT-Brasil) para superar estes aspectos e apoiar o diagnóstico, planejamento e monitoramento de casos endodônticos. Este método de imagem fornece imagens em alta resolução devido a tamanhos submilimétricos de voxel, navegação dinâmica de imagens em vários planos e capacidade de alterar os parâmetros de volume como espessura de corte, intervalos de corte, correção de dados por meio de filtros de imagem, e manipulação do brilho e do contraste. As principais diferenças entre o e-Vol DX e outros software são: compatibilidade com todos os scanners de TCFC atuais com capacidade de exportar dados DICOM, com ajuste de brilho e contraste mais abrangente comparado a outros aplicativos, em que os ajustes são limitados, e geralmente não suportam todos os recursos da faixa dinâmica DICOM; ajuste de espessura de corte personalizado, muitas vezes limitado e pré-definido em outras aplicações; ajuste de nitidez personalizado, muitas vezes limitado em outras aplicações; algoritmo avançado de redução de ruído que melhora a qualidade da imagem; filtros de imagem predefinidos, filtros de para análise de volume do canal radicular com a capacidade de ampliar a imagem em mais de 1000x (reconstruções em 3D) sem perda de resolução, e personalização de parâmetros de imagem automática para melhor padronização e oportunidades de pesquisa; captura com resolução da tela de 192 dpi, com uma opção de 384 dpi, em contraste com os 96 dpi das aplicações similares. Este novo software de TCFC pode apoiar as tomadas de decisões para o tratamento de casos endodônticos complexos e melhorar os resultados do diagnóstico e do tratamento. A melhoria efetiva da qualidade da imagem favorece a prescrição e a interpretação racional das imagens de TCFC.

Correlación en la determinación de la edad ósea radiológica mediante el método de Greulich y Pyle versus la evaluación automatizada utilizando el software BoneXpert / Correlation in radiological bone age determination using the Greulich and Pyle method versus automated evaluation using BoneXpert software

Resumen: Objetivo: Determinar el grado de correlación en la valoración de la edad ósea radiológica mediante el método de Greulich y Pyle versus la evaluación automatizada por el programa computacional BoneXpert® entre los años 2013-2016. Material y Método: Estudio de correlación de técnicas diag nósticas de 1500 radiografías de carpo para evaluar la edad ósea, en pacientes menores de 16 años pertenecientes a Clínica Alemana de Santiago. Las radiografías con evaluación de la edad ósea por el Atlas de Greulich y Pyle (GP) por 1 de 7 radiólogos pediatras fueron sometidas al programa BoneX pert (BE) para la evaluación automatizada de la edad ósea. Se tomó 100 casos al azar para un análisis/ re-análisis del método BE, para conocer su precisión. Se analizó el nivel de correlación de las medicio nes por coeficiente de correlación (r de Pearson) y la variabilidad de las mediciones mediante análisis de Bland-Altman. Resultados: Se incluyeron 1.493 casos, se excluyeron 7 por falla en técnica de la radiografía, 922 de sexo femenino (61.8%), mediana de edad cronológica 9.96 años y 11.12 años para los varones (p 0,001). La correlación entre la edad ósea manual GP y la edad ósea automatizada BE entre los lectores varió entre 0,91 y 0,93. El análisis de Bland-Altman indicó una diferencia promedio entre la edad ósea manual y la edad ósea BE de 0,19 años (IC 0,13 a 0,25). En el análisis/re-análisis de 100 casos al azar mediante BoneXpert, la correlación fue de 1,0. Conclusión: El análisis automatizado mediante BoneXpert permite una evaluación estandarizada, de baja variabilidad, y alta concordancia.

Abstract: Objective: To determine the degree of correlation in the radiological bone age assessment using the Greulich and Pyle method versus automated assessment through BoneXpert® software between 2013 and 2016. Material and Method: Correlation study of diagnostic techniques of 1500 carpal X-rays to assess bone age in patients under 16 years of age from Clínica Alemana de Santiago. X-rays with bone age assessment using the Atlas of Greulich and Pyle (GP) by 1 out of 7 pediatric radiologists, were analyzed using the BoneXpert (BE) software for automated bone age assessment. 100 cases were taken at random for analysis/re-analysis using the BoneXpert method to determine its accuracy. The level of correlation of the measurements was analyzed using the correlation coefficient (Pearson's r) and the variability of the measurements using the Bland-Altman analysis. Results: 1493 cases were assessed, seven were excluded due to failure in the X-ray technique, 922 females (61.8%), with a median chronological age of 9.96 years and 11.12 years for males (p 0.001). The correlation between manual bone age (GP) and automated bone age using BoneXpert method among radiologists ran ged from 0.91 to 0.93. The Bland-Altman analysis indicated an average difference between manual bone age and bone age using the BoneXpert method of 0.19 years (CI 0.13 to 0.25). In the analysis/ re-analysis of 100 random cases using the BoneXpert software, the correlation was 1.00 (100% accu racy). Conclusion: The automated analysis using BoneXpert allows for standardized, low-variability, and high-concordance assessment.