Tongue image feature correlation analysis in benign lung nodules and lung cancer

Objective To analyze the differences in the correlation of tongue image indicators among patients with benign lung nodules and lung cancer.Methods From July 1;2020 to March 31;2022;clinical information of lung cancer patients and benign lung nodules patients was collected at the Oncology Department of Longhua Hos-pital Affiliated to Shanghai University of Traditional Chinese Medicine and the Physical Ex-amination Center of Shuguang Hospital Affiliated to Shanghai University of Traditional Chi-nese Medicine;respectively.We obtained tongue images from patients with benign lung nod-ules and lung cancer using the TFDA-1 digital tongue diagnosis instrument;and analyzed these images with the TDAS V2.0 software.The extracted indicators included color space pa-rameters in the Lab system for both the tongue body(TB)and tongue coating(TC)(TB/TC-L;TB/TC-a;and TB/TC-b);textural parameters[TB/TC-contrast(CON);TB/TC-angular second moment(ASM);TB/TC-entropy(ENT);and TB/TC-MEAN];as well as TC parameters(perAll and perPart).The bivariate correlation of TB and TC features was analyzed using Pearson’s or Spearman’s correlation analysis;and the overall correlation was analyzed using canonical correlation analysis(CCA).Results Samples from 307 patients with benign lung nodules and 276 lung cancer patients were included after excluding outliers and extreme values.Simple correlation analysis indi-cated that the correlation of TB-L with TC-L;TB-b with TC-b;and TB-b with perAll in lung cancer group was higher than that in benign nodules group.Moreover;the correlation of TB-a with TC-a;TB-a with perAll;and the texture parameters of the TB(TB-CON;TB-ASM;TB-ENT;and TB-MEAN)with the texture parameters of the TC(TC-CON;TC-ASM;TC-ENT;and TC-MEAN)in benign nodules group was higher than lung cancer group.CCA further demon-strated a strong correlation between the TB and TC parameters in lung cancer group;with the first and second pairs of typical variables in benign nodules and lung cancer groups indicat-ing correlation coefficients of 0.918 and 0.817(P<0.05);and 0.940 and 0.822(P<0.05);re-spectively.Conclusion Benign lung nodules and lung cancer patients exhibited differences in correla-tion in the L;a;and b values of the TB and TC;as well as the perAll value of the TC;and the texture parameters(TB/TC-CON;TB/TC-ASM;TB/TC-ENT;and TB/TC-MEAN)between the TB and TC.Additionally;there were differences in the overall correlation of the TB and TC be-tween the two groups.Objective tongue diagnosis indicators can effectively assist in the diag-nosis of benign lung nodules and lung cancer;thereby providing a scientific basis for the ear-ly detection;diagnosis;and treatment of lung cancer.

A Survey of Lung Nodules Detection and Classification from CT Scan Images

In the contemporary era,the death rate is increasing due to lung cancer.However,technology is continuously enhancing the quality of well-being.To improve the survival rate,radiologists rely on Computed Tomography(CT)scans for early detection and diagnosis of lung nodules.This paper presented a detailed,systematic review of several identification and categorization techniques for lung nodules.The analysis of the report explored the challenges,advancements,and future opinions in computer-aided diagnosis CAD systems for detecting and classifying lung nodules employing the deep learning(DL)algorithm.The findings also highlighted the usefulness of DL networks,especially convolutional neural networks(CNNs)in elevating sensitivity,accuracy,and specificity as well as overcoming false positives in the initial stages of lung cancer detection.This paper further presented the integral nodule classification stage,which stressed the importance of differentiating between benign and malignant nodules for initial cancer diagnosis.Moreover,the findings presented a comprehensive analysis of multiple techniques and studies for nodule classification,highlighting the evolution of methodologies from conventional machine learning(ML)classifiers to transfer learning and integrated CNNs.Interestingly,while accepting the strides formed by CAD systems,the review addressed persistent challenges.

Evaluation of the dual vascular supply patterns in ground-glass nodules with a dynamic volume computed tomography

BACKGROUND In recent years,the detection rate of ground-glass nodules(GGNs)has been improved dramatically due to the popularization of low-dose computed tomography(CT)screening with high-resolution CT technique.This presents challenges for the characterization and management of the GGNs,which depends on a thorough investigation and sufficient diagnostic knowledge of the GGNs.In most diagnostic studies of the GGNs,morphological manifestations are used to differentiate benignancy and malignancy.In contrast,few studies are dedicated to the assessment of the hemodynamics,i.e.,perfusion parameters of the GGNs.AIM To assess the dual vascular supply patterns of GGNs on different histopathology and opacities.METHODS Forty-seven GGNs from 47 patients were prospectively included and underwent the dynamic volume CT.Histopathologic diagnoses were obtained within two weeks after the CT examination.Blood flow from the bronchial artery[bronchial flow(BF)]and pulmonary artery[pulmonary flow(PF)]as well as the perfusion index(PI)=[PF/(PF+BF)]were obtained using first-pass dual-input CT perfusion analysis and compared respectively between different histopathology and lesion types(pure or mixed GGNs)and correlated with the attenuation values of the lesions using one-way ANOVA,student’s t test and Pearson correlation analysis.RESULTS Of the 47 GGNs(mean diameter,8.17 mm;range,5.3-12.7 mm),30(64%)were carcinoma,6(13%)were atypical adenomatous hyperplasia and 11(23%)were organizing pneumonia.All perfusion parameters(BF,PF and PI)demonstrated no significant difference among the three conditions(all P>0.05).The PFs were higher than the BFs in all the three conditions(all P<0.001).Of the 30 GGN carcinomas,14 showed mixed GGNs and 16 pure GGNs with a higher PI in the latter(P<0.01).Of the 17 benign GGNs,4 showed mixed GGNs and 13 pure GGNs with no significant difference of the PI between the GGN types(P=0.21).A negative correlation(r=-0.76,P<0.001)was demonstrated between the CT attenuation values and the PIs in the 30 GGN carcinomas.CONCLUSION The GGNs are perfused dominantly by the PF regardless of its histopathology while the weight of the BF in the GGN carcinomas increases gradually during the progress of its opacification.

Primary pulmonary meningioma and minute pulmonary meningothelial-like nodules: Rare pulmonary nodular lesions requiring more awareness in clinical practice

In this editorial,we comment on an article by Ruan et al published in a recent issue of the World Journal of Clinical Case.Pulmonary meningothelial proliferative lesions,including primary pulmonary meningiomas,minute pulmonary meningothelial-like nodules,and metastatic pulmonary meningiomas are rare pulmonary lesions.These lesions are difficult to differentiate from lung cancers based on clinical and imaging manifestations.Herein,we briefly introduce the clinical,imaging,and pathological characteristics of these lesions and discuss their pathogenesis to strengthen the current understanding of pulmonary meningothelial proliferative lesions in clinical diagnosis and therapy.

Multi-View Auxiliary Diagnosis Algorithm for Lung Nodules

Lung is an important organ of human body.More and more people are suffering from lung diseases due to air pollution.These diseases are usually highly infectious.Such as lung tuberculosis,novel coronavirus COVID-19,etc.Lung nodule is a kind of high-density globular lesion in the lung.Physicians need to spend a lot of time and energy to observe the computed tomography image sequences to make a diagnosis,which is inefficient.For this reason,the use of computer-assisted diagnosis of lung nodules has become the current main trend.In the process of computer-aided diagnosis,how to reduce the false positive rate while ensuring a low missed detection rate is a difficulty and focus of current research.To solve this problem,we propose a three-dimensional optimization model to achieve the extraction of suspected regions,improve the traditional deep belief network,and to modify the dispersion matrix between classes.We construct a multi-view model,fuse local three-dimensional information into two-dimensional images,and thereby to reduce the complexity of the algorithm.And alleviate the problem of unbalanced training caused by only a small number of positive samples.Experiments show that the false positive rate of the algorithm proposed in this paper is as low as 12%,which is in line with clinical application standards.

Role of inflammatory markers in the evaluation of indeterminate pulmonary nodules:a review

The global mortality rate of lung cancer surpasses that of any other type of cancer,establishing it as the foremost cause of cancer-related deaths worldwide.The objectives of lung nodule assessment are to expedite the diagnosis and treatment of patients with malignant nodules while minimizing unnecessary diagnostic procedures for those with benign nodules.The systemic inflammatory response is closely linked to tumorigenesis.Serum levels of inflammatory markers and their derived parameters markers including neutrophil-to-lymphocyte ratio(NLR),lymphocyte-to-monocyte ratio(LMR),monocyte-to-albumin ratio(MAR),platelet-to-lymphocyte ratio(PLR)and systemic immune‐inflammation index(SII),closely associated with lung cancer.The objective of this article is to comprehensively evaluate the differentiation between benign and malignant pulmonary nodules in terms of inflammatory response indicators,aim at provide practical recommendations for the clinical diagnosis,treatment,and management of pulmonary nodules.

Enhanced characterization of solid solitary pulmonary nodules with Bayesian analysis-based computer-aided diagnosis

The aim of this study was to prospectively assess the accuracy gain of Bayesian analysis-based computeraided diagnosis(CAD) vs human judgment alone in characterizing solitary pulmonary nodules(SPNs) at computed tomography(CT). The study included 100 randomly selected SPNs with a definitive diagnosis. Nodule features at first and follow-up CT scans as well as clinical data were evaluated individually on a 1 to 5 points risk chart by 7 radiologists, firstly blinded then aware of Bayesian Inference Malignancy Calculator(BIMC) model predictions. Raters' predictions were evaluated by means of receiver operating characteristic(ROC) curve analysis and decision analysis. Overall ROC area under the curve was 0.758 before and 0.803 after the disclosure of CAD predictions(P = 0.003). A net gain in diagnostic accuracy was found in 6 out of 7 readers. Mean risk class of benign nodules dropped from 2.48 to 2.29, while mean risk class of malignancies rose from 3.66 to 3.92. Awareness of CAD predictions also determined a significant drop on mean indeterminate SPNs(15 vs 23.86 SPNs) and raised the mean number of correct and confident diagnoses(mean 39.57 vs 25.71 SPNs). This study provides evidence supporting the integration of the Bayesian analysis-based BIMC model in SPN characterization.

Clinical characteristics and work-up of small to intermediate-sized pulmonary nodules in a Chinese dedicated cancer hospital

Objectives:To evaluate the characteristics and work-up of small to intermediate-sized pulmonary nodules in a Chinese dedicated cancer hospital.Methods:Patients with pulmonary nodules 4–25 mm in diameter detected via computed tomography(CT)in 2013 were consecutively included.The analysis was restricted to patients with a histological nodule diagnosis or a 2-year follow-up period without nodule growth confirming benign disease.Patient information was collected from hospital records.Results:Among the 314 nodules examined in 299 patients,212(67.5%)nodules in 206(68.9%)patients were malignant.Compared to benign nodules,malignant nodules were larger(18.0 mm vs.12.5 mm,P<0.001),more often partly solid(16.0%vs.4.7%,P<0.001)and more often spiculated(72.2%vs.41.2%,P<0.001),with higher density in contrast-enhanced CT(67.0 HU vs.57.5 HU,P=0.015).Final diagnosis was based on surgery in 232 out of 314(73.9%)nodules,166 of which were identified as malignant[30(18.1%)stage III or IV]and 66 as benign.In 36 nodules(11.5%),diagnosis was confirmed by biopsy and the remainder verified based on stability of nodule size at follow-up imaging(n=46,14.6%).Among 65 nodules subjected to gene(EGFR)mutation analyses,28(43.1%)cases(EGFR19 n=13;EGFR21 n=15)were identified as EGFR mutant and 37(56.9%)as EGFR wild-type.Prior to surgery,the majority of patients[n=194(83.6%)]received a contrast-enhanced CT scan for staging of both malignant[n=140(84.3%)]and benign[n=54(81.8%)]nodules.Usage of positron emission tomography(PET)-CT was relatively uncommon[n=38(16.4%)].Conclusions:CT-derived nodule assessment assists in diagnosis of small to intermediate-sized malignant pulmonary nodules.Currently,contrast-enhanced CT is commonly used as the sole diagnostic confirmation technique for pre-surgical staging,often resulting in surgery for late-stage disease and unnecessary surgery in cases of benign nodules.

Pathological diagnosis and immunohistochemical analysis of minute pulmonary meningothelial-like nodules:A case report

BACKGROUND Minute Pulmonary Meningothelial-like Nodules(MPMNs)are rare benign pulmonary nodules,which are more common in elderly women and have a higher detection rate in lung tissues of patients with lung malignant diseases.Its origin is not yet clear.At present,there are few reports on the diagnostic methods such as imaging and pathological manifestations of MPMNs.This article reports a 70-year-old female patient with pulmonary adenocarcinoma combined with MPMNs and reviews of the relevant literature.CASE SUMMARY A 70-year-old women was admitted to our institution with feeling sour in her back and occasional cough for more than 2 mo.Computerized electronic scanning scan and 3D reconstruction images in our institution showed there were multiple ground-glass nodules in both of her two lungs.The biggest one was in the apicoposterior segment of left upper lobe,about 2.5 mm×9 mm in size.We performed thoracoscopic resection of the left upper lung apicoposterior segment of the patient,and the final pathological report was minimally invasive adenocarcinoma.Re-examination of high resolution computed tomography 21 mo after surgery showed multiple ground-glass nodules in both lungs,and a new groundglass nodule was found in the superior segment of the right lower lobe.We took pathological biopsy of the right upper lung and right lower lung nodules for the patient under thoracoscopy.The histomorphology of the right lower lobe nodule showed multiple lesions in the lung tissue,and the small foci in the alveolar septum were distributed in mild form of the aggregation of short spindle cells.The immunohistochemistry showed that the lesion was epithelial membrane antigen(EMA)(+),somatostatin receptor 2a(SSTR2a)(+),S-100(-),chromogranin A(-),Syn(-),cytokeratin(-)and HMB-45(-).The final diagnosis was minimally invasive adenocarcinoma,accompanied by MPMNs.We recommend that patients continue to receive treatment after surgery and to do regular follow-up observations.CONCLUSION The imaging manifestations of MPMNs are atypical,histomorphology and immunohistochemistry can assist in its diagnosis.This article reviews the relevant literature of MPMNs immunohistochemistry and shows that MPMNs are positive for EMA,SSTR2a,and progesterone receptor.

Computed tomography fluoroscopy guided percutaneous lung biopsy for ground-glass opacity pulmonary lesions:A meta-analysis

AIM： To obtain the diagnostic performance of percu-taneous transthoracic needle biopsy （PTNB） under Computed tomography （CT） fuoroscopy guidance for lung ground-glass opacity （GGO）.METHODS： We searched for English- and Chinese-language studies in PubMed, EMBASE, EBSCO, OVID, and CNKI （China National Knowledge Infrastructure） database. Data were calculated with Meta-Disc version 1.4 and Rev Man version 5.2 software. From the pooled data, we calculated sensitivity （Sen）, specificity （Spe）, positive likelihood ratio （＋LR）, negative likelihood ratio （-LR）, and diagnostic odds ratio （DOR）. Summary receiver operating characteristic （SROC） curves were constructed and incidence of complications was recorded.RESULTS： Four documents included in this present meta-analysis met the criteria for analysis. The pooled Sen, Spe, ＋LR, -LR and DOR with 95%CI were 0.91 （0.86-0.95）, 1.0 （0.91-1.0）, 18.64 （4.83-71.93）, 0.11 （0.05-0.26） and 153.17 （30.78-762.33）, respectively. The area under the SROC curve was 0.98. The incidence of pneumothorax and hemoptysis was 17.86%-51.80% and 10.50%-19.40%, respectively.CONCLUSION： CT fuoroscopy-guided PTNB, which has an acceptable incidence of complications, can be used as a primary examination method for lung GGO, with moderate sensitivity and specifcity.

Feasible Method to Assess the Performance of a Lung Cancer CT Screening CAD System in Clinical Practice: Dependence on Nodule Size and Density

Detection of small pulmonary nodules is the goal of lung cancer screening. Computer-aided detection (CAD) systems are recommended to use in lung cancer computed tomography (CT) screening to increase the accuracy of nodule detection. Size and density of lung nodules are primary factors in determining the risk of malignancy. Therefore, purpose of this study is to apply computer-simulated virtual nodules based on the point spread function (PSF) measured in same scanner (maintaining spatial resolution condition) to assess the CAD system performance dependence on nodule size and density. Virtual nodules with density differences between lung background and nodule density (ΔCT) values (200, 300 and 400 HU) and different sizes (4 to 8 mm) were generated and fused on clinical images. CAD detection was performed and free-response receiver operating characteristic (FROC) curves were obtained. Results show that both density and size of virtual nodules can affect detection efficiency. Detailed results are possible to use for quantitative analysis of a CAD system performance. This study suggests that PSF-based virtual nodules could be effectively used to assess the lung cancer CT screening CAD system performance dependence on nodule size and density.

CTCs Detection and Whole-exome Sequencing Might Be Used to Differentiate Benign and Malignant Pulmonary Nodules

Background and objective Low-density computed tomography(LDCT)improved early lung cancer diagnosis but introduces an excess of false-positive pulmonary nodules data.Hence,accurate diagnosis of early-stage lung cancer remains challenging.The purpose of the study was to assess the feasibility of using circulating tumour cells(CTCs)to differentiate malignant from benign pulmonary nodules.Materials and methods 122 patients with suspected malignant pulmonary nodules detected on chest CT in preparation for surgery were prospectively recruited.Peripheral blood samples were collected before surgery,and CTCs were identified upon isolation by size of epithelial tumour cells and morphological analysis.Laser capture microdissection,MALBAC amplification,and whole-exome sequencing were performed on 8 samples.The diagnostic efficacy of CTCs counting,and the genomic variation profile of benign and malignant CTCs samples were analysed.Results Using 2.5 cells/5 m L as the cut-off value,the area under the receiver operating characteristic curve was of 0.651(95%confidence interval:0.538-0.764),with a sensitivity and specificity of 0.526 and 0.800,respectively,and positive and negative predictive values of 91.1%and 30.3%,respectively.Distinct sequence variations differences in DNA damage repair-related and driver genes were observed in benign and malignant samples.TP53 mutations were identified in CTCs of four malignant cases;in particular,g.7578115T>C,g.7578645C>T,and g.7579472G>C were exclusively detected in all four malignant samples.Conclusion CTCs play an ancillary role in the diagnosis of pulmonary nodules.TP53 mutations in CTCs might be used to identify benign and malignant pulmonary nodules.

Centralized Lung Nodule Management at A VeteransHospital Using A Multidisciplinary Lung NoduleEValuation Team(LNET)

Introduction: Lung nodules are frequently identified on imaging studies and can represent early lung cancers. We instituted the Lung Nodule Evaluation Team(LNET) to optimize management of these nodules by a lung specialist physician. All lung nodules identified by a radiologist prompted a direct consultation to this service. We report our initial experience with this process. Methods: This is a retrospective review of patients with lung nodules at a single institution from 2008 to 2015. Since October 2014, lung nodules >3 mm identified on computed tomography(CT) scanning of the chest generate an automatic consult to LNET from the radiology service. Demographic, nodule and follow up data was entered into a surveillance database and summarized.Results: There were 1,873 patients identified in the database. Of these, 900 patients were undergoing active surveillance. Consults increased from 5.5 to 93 per month after the start of the new consult program. Lung nodules were identified on 64% of chest CT scans. Prior to the direct radiology consult the average size of a nodule was 1.7 cm and 0.7 cm after. The overall time from initial nodule imaging to initiating a management plan by a thoracic specialist physician was 3.7 days. Conclusion: Assessment of lung nodules by a specialist physician is important to ensure appropriate long term management and optimize utilization of diagnostic interventions. A direct radiology consult to a specialized team of chest physicians decreased the time in initiating a management plan, identified smaller nodules and may lead to a more judicious use of health care resources in the management of lung nodules.

Lung Nodule Image Retrieval Based on Convolutional Neural Networks and Hashing

Lung medical image retrieval based on content similarity plays an important role in computer-aided diagnosis of lung cancer.In recent years,binary hashing has become a hot topic in this field due to its compressed storage and fast query speed.Traditional hashing methods often rely on highdimensional features based hand-crafted methods,which might not be optimally compatible with lung nodule images.Also,different hashing bits contribute to the image retrieval differently,and therefore treating the hashing bits equally affects the retrieval accuracy.Hence,an image retrieval method of lung nodule images is proposed with the basis on convolutional neural networks and hashing.First,apre-trained and fine-tuned convolutional neural network is employed to learn multilevel semantic features of the lung nodules.Principal components analysis is utilized to remove redundant information and preserve informative semantic features of the lung nodules.Second,the proposed method relies on nine sign labels of lung nodules for the training set,and the semantic feature is combined to construct hashing functions.Finally,returned lung nodule images can be easily ranked with the query-adaptive search method based on weighted Hamming distance.Extensive experiments and evaluations on the dataset demonstrate that the proposed method can significantly improve the expression ability of lung nodule images,which further validates the effectiveness of the proposed method.

Evaluating the Growth of Pulmonary Nodular Ground-glass Opacity on CT: Comparison of Volume Rendering and Thin Slice Images

This study examined the value of volume rendering （VR） interpretation in assessing the growth of pulmonary nodular ground-glass opacity （nGGO）. A total of 47 nGGOs （average size, 9.5 mm; range, 5.7-20.6 mm） were observed by CT scanning at different time under identical parameter settings. The growth of nGGO was analyzed by three radiologists by comparing the thin slice （TS） CT images of initial and repeat scans with side-by-side cine mode. One week later synchronized VR images of the two scans were compared by side-by-side cine mode to evaluate the nGGO growth. The nodule growth was rated on a 5-degree scale： notable growth, slight growth, dubious growth, stagnant growth, shrinkage. Growth standard was defined as： Density increase 〉 30 HU and （or） diameter increase （by 20% in nodules 〉_10 mm, 30% in nodules of 5-9 mm）. Receiver operating characteristic （ROC） was performed. The results showed that 32 nGGOs met the growth criteria （29 nGGOs showed an increase in density; 1 nGGO showed an increase in diameter; 2 nGGOs showed an increase in both diameter and density）. Area under ROC curve revealed that the performance with VR interpretation was better than that with TS interpretation （P〈0.01, P〈0.05 and P〈0.05 for observers A, B and C respectively）. Consistency between different observers was excellent with both VR interpretation （κ=0.89 for observers A＆C A＆B, B＆C） and TS interpretation （κ=0.71 for A＆B, κ=0.68 for A＆C, κ=0.74 for B＆C）, but time spending was less with VR interpretation than with TS interpretation （P〈0.0001, P〈0.0001 and P〈0.05 for observers A, B and C, respectively）. It was concluded that VR is a useful technique for evaluating the growth of nGGO.

Clinical and Pathological Research Status of Multiple Pulmonary Nodules

With the changes in disease spectrum and the popularization of screening of low-dose spiral CT (CT) in the chest, more and more pulmonary nodules have been detected, most of which are bipulmonary multiple nodules. The existence of multiple pulmonary nodules means that it may be a pathological state of benign and malignant co-existence. The origin and evolution of pulmonary nodules in different histopathological states have a great impact on the choice of treatment methods. In recent years, the rise of immunotherapy has brought a breakthrough in the treatment of refractory lung cancer. However, some patients are still ineffective in immunotherapy, which may be related to the immune microenvironment where nodules are proportioned in different components in different pathological states. This review article mainly predicts the development process of nodules by analyzing the origin of multiple pulmonary nodules and the immune microenvironment of nodules in different pathological conditions, so as to provide guidance for clinical treatment.

End-to-End 2D Convolutional Neural Network Architecture for Lung Nodule Identification and Abnormal Detection in Cloud

The extent of the peril associated with cancer can be perceivedfrom the lack of treatment, ineffective early diagnosis techniques, and mostimportantly its fatality rate. Globally, cancer is the second leading cause ofdeath and among over a hundred types of cancer;lung cancer is the secondmost common type of cancer as well as the leading cause of cancer-relateddeaths. Anyhow, an accurate lung cancer diagnosis in a timely manner canelevate the likelihood of survival by a noticeable margin and medical imagingis a prevalent manner of cancer diagnosis since it is easily accessible to peoplearound the globe. Nonetheless, this is not eminently efficacious consideringhuman inspection of medical images can yield a high false positive rate. Ineffectiveand inefficient diagnosis is a crucial reason for such a high mortalityrate for this malady. However, the conspicuous advancements in deep learningand artificial intelligence have stimulated the development of exceedinglyprecise diagnosis systems. The development and performance of these systemsrely prominently on the data that is used to train these systems. A standardproblem witnessed in publicly available medical image datasets is the severeimbalance of data between different classes. This grave imbalance of data canmake a deep learning model biased towards the dominant class and unableto generalize. This study aims to present an end-to-end convolutional neuralnetwork that can accurately differentiate lung nodules from non-nodules andreduce the false positive rate to a bare minimum. To tackle the problem ofdata imbalance, we oversampled the data by transforming available images inthe minority class. The average false positive rate in the proposed method isa mere 1.5 percent. However, the average false negative rate is 31.76 percent.The proposed neural network has 68.66 percent sensitivity and 98.42 percentspecificity.

Attention Based Multi-Patched 3D-CNNs with Hybrid Fusion Architecture for Reducing False Positives during Lung Nodule Detection

In lung nodules there is a huge variation in structural properties like Shape, Surface Texture. Even the spatial properties vary, where they can be found attached to lung walls, blood vessels in complex non-homogenous lung structures. Moreover, the nodules are of small size at their early stage of development. This poses a serious challenge to develop a Computer aided diagnosis (CAD) system with better false positive reduction. Hence, to reduce the false positives per scan and to deal with the challenges mentioned, this paper proposes a set of three diverse 3D Attention based CNN architectures (3D ACNN) whose predictions on given low dose Volumetric Computed Tomography (CT) scans are fused to achieve more effective and reliable results. Attention mechanism is employed to selectively concentrate/weigh more on nodule specific features and less weight age over other irrelevant features. By using this attention based mechanism in CNN unlike traditional methods there was a significant gain in the classification performance. Contextual dependencies are also taken into account by giving three patches of different sizes surrounding the nodule as input to the ACNN architectures. The system is trained and validated using a publicly available LUNA16 dataset in a 10 fold cross validation approach where a competition performance metric (CPM) score of 0.931 is achieved. The experimental results demonstrate that either a single patch or a single architecture in a one-to-one fashion that is adopted in earlier methods cannot achieve a better performance and signifies the necessity of fusing different multi patched architectures. Though the proposed system is mainly designed for pulmonary nodule detection it can be easily extended to classification tasks of any other 3D medical diagnostic computed tomography images where there is a huge variation and uncertainty in classification.

Dual time point FDG PET imaging in evaluating pulmonary nodules with low FDG avidity

A standardized uptake value(SUV) of 2.5 is frequently used as criteria to evaluate pulmonary lesions.However,false results may occur.Some studies have shown the usefulness of delayed PET for improving accuracy,while others recently have shown fewer promising results.This study was designed to investigate the accuracy of dual time point(DTP) FDG PET imaging in the evaluation of pulmonary lesions with an initial SUV less than 2.5.DTP FDG PET studies were conducted about 1 and 2 hours after FDG injection,and pulmonary lesions with an initial SUV less than 2.5 were identified.Nodules with pathologic results or imaging follow up were included.The differences in SUV and retention index(RI) between benign and malignant pulmonary lesions were analyzed.Receiver operating characteristics(ROC) analysis was performed to evaluate the discriminating validity of SUV and RI.51 lesions were finally included.A RI greater than 0% was observed in 64% of the benign lesions;56% had a RI greater than 10%.Among the malignancies,80.8% had a RI greater than 0%,and 61.5% had a RI greater than 10%.We found no significant differences in SUV and RI between benign and malignant lesions.The area under the ROC curve did not differ from 0.5 whether using SUV or the retention index.Utilizing a SUV increase of 10%,the sensitivity was 61.5%,specificity 44% and accuracy was 52.9%.Dual time point FDG PET may not be of benefit in the evaluation of pulmonary nodules with low FDG avidity.

INCREMENTAL DYNAMIC COMPUTER TOMOGRAPHY IN SURGICAL TREATMENT OF PATIENTS WITH SOLITARY PULMONARY NODULES

Objective The purpose of this study is to appraise the value of incremental dynamic enhanced computed tomography in surgical treatment of patients with solitary pulmonary nodules(SPNs).Methods The data of 42 cases with solitary pulmonary nodules who underwent surgical treatment from May 2002 to June 2003 in our hospital were collected to find the relationship between preoperative dynamic enhanced CT image and postoperative pathology.Result All bronchogenic carcinoma showed significant enhancement after intravenous 100 mL iodinated contrast material.The average degree of enhancement of bronchogenic carcinoma was significantly different from that of tuberculoma and other benign lesions.Conclusion Dynamic enhanced computed tomography is very valuable in distinguishing between malignant nodules and benign ones.Emphasis should be paid to lymph nodes in the dynamic enhanced computed tomography,which is useful both to the diagnosis of SPN and for surgical treatment.