Detectability of lung nodules using flat panel detector with dual energy subtraction by two shot method: Evaluation by ROC method

We retrospectively evaluated the usefulness of computer-aided diagnosis (CAD) schemes to radiologist performance in the simultaneous detection of vertebral fractures and lung nodules on chest radiographs. We evaluated posteroanterior and lateral chest images of 21 patients with vertebral fractures, 31 patients with lung nodules, and 10 persons acting as controls. The total number of subjects was 60 because both lesions were present in four patients. Eighteen radiologists were asked to detect vertebral fractures and nodules simultaneously on posteroanterior and lateral images. The radiologists indicated their confidence level ratings regarding the presence or absence of lesions and the most likely location of each lesion on either posteroanterior or lateral images, first without and then with CAD output. The observers' performance was evaluated with use of receiver operating characteristic (ROC) and jackknife free-response ROC curves. With the CAD scheme, the average area under the ROC curve for detection of vertebral fractures improved from 0.906 to 0.951 (p = 0.002). That for lung nodules also improved, but the improvement was not statistically significant (0.804-0.816, p = 0.297). The figure-of-merit values obtained with the jackknife free-response ROC program improved from 0.585 to 0.680 (p < 0.001) for vertebral fractures and from 0.622 to 0.650 (p = 0.017) for nodules, both results having statistical significance. Average sensitivity in the detection of lesions improved from 59.8% to 69.3% for vertebral fractures and from 64.9% to 67.6% for nodules. In the detection of vertebral fractures and lung nodules on chest images, diagnostic accuracy among radiologists improves with the use of CAD.

A survey and taxonomy of 2.5D approaches for lung segmentation and nodule detection in CT images

Detection of small pulmonary nodules on chest radiographs: efficacy of dual-energy subtraction technique using flat-panel detector chest radiography

To determine the value of digital storage-phosphor radiography (SR) on the detection and identification of subtle lung nodules, postero-anterior (PA) and lateral (LAT) film-screen (FR) chest radiographs were compared with isodose SR images of 45 patients with metastatic malignancies. The SR postprocessing was done with a particular mode previously optimized for routine chest radiography. Pulmonary metastases were found in 34 patients and were proved or excluded by CT (n = 28) or longterm follow-up FR (n = 17). Chest images were divided into four regions for evaluation of image quality, number of lung nodules per region and marked pulmonary structures by receiver-operating characteristics (ROC) analysis (45 patients; 125 nodules; 2810 observations; five readers). Of the nodules selected for an ROC study 82 % were 0.5–1.0 cm in diameter. Overall image quality was rated better for FR concerning lung fields (PA) and mediastinum/hilum (LAT). More lung-field nodules were detected on FR than on SR chest images (P < 0.05). Use of FR was superior to SR in the general identification of nodules (PA chest), especially concerning intermediate (P < 0.01) and subtle abnormalities (P < 0.05), whereas there was no significant difference for LAT chest images. Our results show, that currently FR still has advantages over SR in the detection and identification of subtle lung nodules in routine clinical radiography.

To determine whether use of bone suppression (BS) imaging, used together with a standard radiograph, could improve radiologists' performance for detection of small lung cancers compared with use of standard chest radiographs alone and whether BS imaging would provide accuracy equivalent to that of dual-energy subtraction (DES) radiography. Institutional review board approval was obtained. The requirement for informed consent was waived. The study was HIPAA compliant. Standard and DES chest radiographs of 50 patients with 55 confirmed primary nodular cancers (mean diameter, 20 mm) as well as 30 patients without cancers were included in the observer study. A new BS imaging processing system that can suppress the conspicuity of bones was applied to the standard radiographs to create corresponding BS images. Ten observers, including six experienced radiologists and four radiology residents, indicated their confidence levels regarding the presence or absence of a lung cancer for each lung, first by using a standard image, then a BS image, and finally DES soft-tissue and bone images. Receiver operating characteristic (ROC) analysis was used to evaluate observer performance. The average area under the ROC curve (AUC) for all observers was significantly improved from 0.807 to 0.867 with BS imaging and to 0.916 with DES (both P < .001). The average AUC for the six experienced radiologists was significantly improved from 0.846 with standard images to 0.894 with BS images (P < .001) and from 0.894 to 0.945 with DES images (P = .001). Use of BS imaging together with a standard radiograph can improve radiologists' accuracy for detection of small lung cancers on chest radiographs. Further improvements can be achieved by use of DES radiography but with the requirement for special equipment and a potential small increase in radiation dose.

The objective of our study was to compare the effect of dual-energy subtraction and bone suppression software alone and in combination with computer-aided detection (CAD) on the performance of human observers in lung nodule detection. One hundred one patients with from one to five lung nodules measuring 5-29 mm and 42 subjects with no nodules were retrospectively selected and randomized. Three independent radiologists marked suspicious-appearing lesions on the original chest radiographs, dual-energy subtraction images, and bone-suppressed images before and after postprocessing with CAD. Marks of the observers and CAD marks were compared with CT as the reference standard. Data were analyzed using nonparametric tests and the jackknife alternative free-response receiver operating characteristic (JAFROC) method. Using dual-energy subtraction alone (p = 0.0198) or CAD alone (p = 0.0095) improved the detection rate compared with using the original conventional chest radiograph. The combination of bone suppression and CAD provided the highest sensitivity (51.6%) and the original nonenhanced conventional chest radiograph alone provided the lowest (46.9%; p = 0.0049). Dual-energy subtraction and bone suppression provided the same false-positive (p = 0.2702) and true-positive (p = 0.8451) rates. Up to 22.9% of lesions were found only by the CAD program and were missed by the readers. JAFROC showed no difference in the performance between modalities (p = 0.2742-0.5442). Dual-energy subtraction and the electronic bone suppression program used in this study provided similar detection rates for pulmonary nodules. Additionally, CAD alone or combined with bone suppression can significantly improve the sensitivity of human observers for pulmonary nodule detection.

The lung nodule detection technology plays a vital role in the diagnosis and treatment of early lung cancer. Deep learning is currently one of the main technologies applied in computer vision related fields. Therefore, this topic is to combine deep learning and lung nodule detection technology. Different from some research methods, we improved from two-dimensional space to three- dimensional space and applied it to lung nodule detection. After statistically analyzing the lung CT data, we find that the lung CT data has inconsistent scales on the vertical axis. Aiming at the problem of lung CT data, this article is based on the deep learning method. First, the lung interpolation network based on voxel flow is used to achieve the same scale, and then the lung nodule detection network based on three-dimensional convolution is used to complete the detection of lung nodules. The entire network combines U- Net-like and RPN-like network structures. Through data slice input, it avoids the limitation of the display memory of the computing platform. The network structure also introduces prior knowledge of the coordinate information of the input slices to improve the classification accuracy of lung nodules. The experimental results show that the lung slice data of consistent scale is achieved through interpolation, and then through the three-dimensional convolution of the lung nodule detection network, the state- of-the-art detection effect is achieved. Because of the introduction of the interpolation network, the time- consuming has increased. Of course, the overall speed stillmeets the actual use value.

Comparison of Chest Dual-energy Subtraction Digital Tomosynthesis and Dual-energy Subtraction Radiography for Detection of Pulmonary Nodules: Initial Evaluations in Human Clinical Cases

Detection of lung nodules is key in the treatment of early-stage lung cancer. Computed tomography (CT) scanning technology is an essential contactless tool. However, stray radiation caused by a patient's slight movements and equipment operation can impair CT images, hindering accurate lung nodule detection. To address these issues, this study proposes an artificial intelligence-based anti-interference lung nodule detection method, which is primarily structured with Yolov8 and combines the modules of adaptive gating sparse attention (AGSA) and haar wavelet downsampling (HWD), referred to as Yolov8-AH. This model aimed to improve the accuracy of lung nodule detection in lung CT images under interference conditions. AGSA focuses on key areas of the image, promoting detection stability even when CT images are disturbed. Furthermore, HWD prioritizes the frequency components corresponding to the size and shape of the nodules, enhancing their visibility for easier detection and analysis. HWD effectively reduces image noise without significantly blurring the lung nodule edges, emphasizing them prominently within the lung tissue. Furthermore, when combined with the Yolov8 deep learning model driven by artificial intelligence, the model could accurately detect lung nodules, significantly aiding in early diagnosis and treatment. The effectiveness of the Yolov8-AH detection model was verified through ablation experiments, experiments under varying noise intensities, and experiments under different noise application ratios. The experimental results demonstrate that, compared to existing lung nodule detection models, the Yolov8-AH model achieves a 24% improvement in mAP50 and an 8.2% improvement in precision.

Contrast-enhanced (CE) breast imaging involves the injection contrast agents (i.e., iodine) to increase conspicuity of malignant lesions. CE imaging may be used in conjunction with digital mammography (DM) or digital breast tomosynthesis (DBT) and has shown promise in improving diagnostic specificity. Both CE-DM and CE-DBT techniques require optimization as clinical diagnostic tools. Physical factors including x-ray spectra, subtraction technique, and the signal from iodine contrast, must be considered to provide the greatest object detectability and image quality. We developed a cascaded linear system model (CLSM) for the optimization of CE-DM and CE-DBT employing dual energy (DE) subtraction or temporal (TE) subtraction. We have previously developed a CLSM for DBT implemented with an a-Se flat panel imager (FPI) and filtered backprojection (FBP) reconstruction algorithm. The model is used to track image quality metrics - modulation transfer function (MTF) and noise power spectrum (NPS) - at each stage of the imaging chain. In this study, the CLSM is extended for CE breast imaging. The effect of x-ray spectrum (varied by changing tube potential and the filter) and DE and TE subtraction techniques on breast structural noise was measured was studied and included as a deterministic source of noise in the CLSM. From the two-dimensional (2D) and three-dimensional (3D) MTF and NPS, the ideal observer signal-to-noise ratio (SNR), also known as the detectability index (d'), may be calculated. Using d' as a FOM, we discuss the optimization of CE imaging for the task of iodinated contrast object detection within structured backgrounds. Increasing x-ray energy was determined to decrease the magnitude of structural noise and not its correlation. By performing DE subtraction, the magnitude of the structural noise was further reduced at the expense of increased stochastic (quantum and electronic) noise. TE subtraction exhibited essentially no residual structural noise at the expense of increased quantum noise, even over that of the DE case. For DE subtraction, optimization of dose weighting to the HE view (fh ) results in the minimization of quantum noise. Both subtraction weighting factor (wSub ) and the iodine contrast signal were dependent on the LE and HE x-ray spectra. To best detect a 5 mm Gaussian lesion with 5 mg/ml of iodine within a 4 cm thick breast, it was found that the high energy (HE) view should be acquired with a tube potential of 47 kVp (W/Ti spectrum) and the low energy (LE) view with a potential of 23 kVp (W/Rh spectrum). Due to the complete removal of structural noise, TE subtraction produced much higher d' than DE subtraction both as a function of mean glandular dose and iodine concentration. We have shown the effect of increasing x-ray energy as well as projection domain subtraction on breast structural noise. Further, we have exhibited the utility of the CLSM for DE and TE subtraction CE imaging in the optimization of imaging parameters such as x-ray energy, fh , and wSub as well as guiding the understanding of their effects on image contrast and noise.

Receiver operating characteristic (ROC) methods are useful tools for evaluating the sensitivity and specificity of various postprocessing algorithms used in fMRI data analysis. New ROC methods using real fMRI data are proposed that improve a previously introduced method by Le and Hu (Le and Hu, NMR Biomed 1997;10:160-164). The proposed methods provide more accurate means of estimating the true ROC curve from real data and thereby aid in the comparative evaluation of a wide range of postprocessing tools in fMRI. The mathematical relationships between different ROC curves are explored for a comparison of different ROC methods. Examples using real and simulated data are provided to illustrate the ideas involved.

The objective of many existing computer-aided diagnosis (CADx) schemes for lung nodule detection is to reduce the number of false-positives (i.e., increase specificity) while maintaining a high level of sensitivity. Our examination of the false-positives obtained with the previously developed CADx program show that many round objects, such as rib crossings, end-on vessels, and aggregates of vessels, were mistakenly classified as nodules. Among the problems of decreasing the number of false-positives, the differentiation between nodules and end-on vessels is one of the most challenging tasks performed by computers. To eliminate the false-positives, two methods are proposed. One method is to extract the known features (i.e., contrast and size) based on a conventional digital image processing technique. The other method uses an artificial neural network (ANN) which is specifically trained to classify nodules and end-on vessels. Performances of the two approaches are evaluated using the receiver operating characteristics (ROC) method and the area under the ROC curve (A<SUB>z</SUB>). Based on our test database, the FFNN and the algorithmic approaches showed preliminary ROC performances with A<SUB>z</SUB> values equal to 0.90 and 0.94, respectively.

The purpose of this study was to discuss the optimal X-ray beam quality for detection of simulated lung nodule on a computed radiography system. We set up four types of X-ray beam quality (90 kV, 120 kV, 150 kV, and 120 kV + gadolinium filter), and kept the incident dose on the patient at 0.3 mGy. A receiver operating characteristic (ROC) analysis and a granularity measurement were used to evaluate the relationship between the detection of a low-contrast object and X-ray beam quality. As a result, the areas under the ROC curve (AUC) of 90 kV and 120 kV + gadolinium filter were significantly superior to those of 120 kV and 150 kV (p<0.05). However, a significant difference was not observed between 90 kV and 120 kV + gadolinium filter, 120 kV and 150 kV. The order of the granularity values gave good agreement with the results of visual evaluation. In conclusion, we considered that the optimal X-ray beam quality was 90 kV or 120 kV + gadolinium filter.

Receiver operating characteristic (ROC) analysis has been used in a variety of settings since it was first declassified by the United States government over 60 years ago. One venue in which it has received particular attention is in the field of radiology. In radiology, as in other areas of application, ROC analysis is used to assess the ability of a diagnostic test to distinguish between two opposing states. One useful descriptor in ROC analysis is the area under the ROC curve. At times, it is useful and insightful to average ROC curves in order to create a single curve that summarizes all of the data from multiple readers. In this paper, we investigate the impact of correlated readings on the average area under two readers' ROC curves using several common averaging strategies, and then apply the results to a radiologic study.

A critical step to ameliorate diagnosis and extend patient survival is Benign-malignant Pulmonary Nodule (PN) classification at earlier detection. On account of the noise of Computed Tomography (CT) images, the prevailing Lung Nodule (LN) detection techniques exhibit broad variation in accurate prediction. Thus, a novel Nodule Detection along with Classification algorithm for early diagnosis of Lung Cancer (LC) has been proposed. Initially, employing the Adaptive Mode Ostu Binarization (AMOB) technique, the Lung Volumes (LVs) isextortedas of the image together with the extracted lung regions is pre-processed. Then, detection of LNs takes place, and utilizing Geodesic Fuzzy C-Means Clustering (GFCM) Segmentation Algorithm, it is segmented.Next, the vital features are extracted, and the Nodules are classified by utilizing Logarithmic Layer Xception Neural Network (LLXcepNN) Classifier grounded on the extracted feature. The nodules are classified as Benign Nodules (BN) and Malignant Nodules (MN) by the proposed classifier. Lastly, the Lung CT images are scrutinized. Thus, when weighed against the prevailing techniques, the proposed systems' acquired outcomes exhibit that the rate of accuracy of classification is enhanced.