An evaluation of consensus techniques for diagnostic interpretation

There is considerable research in the field of content-based medical image retrieval; however, few of the current systems investigate the relationship between the radiologists' visual impression of image similarity and the computer calculated content-based similarity. Furthermore, those research studies that investigate these relationships analyze the visual similarity with respect to degree of malignancy without including specific characteristics that are important in the diagnosis process.

Detection of pulmonary nodules with ground glass opacity (GGO) is a difficult task in radiology. Follow up is often required in medical fields. But diagnosis based on CT images are dependent on ability and experience of radiologists. In addition to that, enormous number of images increase their burden. So, to improve the detection accuracy and to reduce the burden of doctors, a CAD (Computer Aided Diagnosis) system is expected. So, in this paper, we propose an automatic algorithm for GGO detection on CT images. At first, vessel areas are removed from original CT images by using 3D Line Filter and then candidate regions are detected by threshold processing. After that, we calculate statistical features of segmented candidate regions and use artificial neural network (ANN) to distinguish final candidate regions. We applied the proposed method to 31 CT image sets in the Lung Image Database Consortium (LIDC) which is supplied by National Center Institute (NCI). In this paper, we show the experimental results and give discussions.

Lung Cancer is one of the most deadly diseases worldwide. According to the American Cancer Society, about 234,030 peoples have been suffering from lung cancer. It can be cured if it is diagnosed earlier which decreases the death rate. A computational diagnostic tool named Computer Aided Diagnosis (CAD) is used to detect pulmonary nodules. Extensive work has been made in this domain. However, previous Computer Aided Diagnosis (CAD) system are time-consuming since they needed more modules such as image modification, segmentation and the features should be extracted by the domain experts to build the entire CAD system. It is hard to examine large data using the existing CAD system. Thus, a novel framework with a Convolutional Neural Network (CNN) to detect pulmonary nodule is proposed. Firstly, a preprocessing technique named bilateral filtering is applied to increase the image quality and remove the irrelevant noise from the Computer Tomography (CT) images. Secondly, the preprocessed data are trained into a convolutional neural network to detect the nodule and classify it. The performance of this system is validated using the Lung Image Database Consortium (LIDC) dataset. The accuracy of nodule candidate detection achieves 93%. It states that the proposed method achieves better accuracy in nodule detection.

Medical imaging technology has always provided radiologists with the opportunity to view and keep records of anatomy of the patient. With the development of machine learning and intelligent computing, these images can be used to create Computer-Aided Diagnosis (CAD) systems, which can assist radiologists in analyzing image data in various ways to provide better health care to patients. This paper looks at increasing accuracy and reducing cost in creating CAD systems, specifically in predicting the malignancy of lung nodules in the Lung Image Database Consortium (LIDC). Much of the cost in creating an accurate CAD system stems from the need for multiple radiologist diagnoses or annotations of each image, since there is rarely a ground truth diagnosis and even different radiologists' diagnoses of the same nodule often disagree. To resolve this issue, this paper outlines an method of selective iterative classification that predicts lung nodule malignancy by using multiple radiologist diagnoses only for cases that can benefit from them. Our method achieved 81% accuracy while costing only 46% of the method that indiscriminately used all annotations, which achieved a lower accuracy of 70%, while costing more.

Computer Aided Detection (CAD) systems provide a second opinion to radiologists in detecting lung cancer by providing automated analysis of the scans. The proposed CAD system consists of five processing steps: image acquisition, preprocessing, lung segmentation, nodule detection and false positive reduction. First, 400 CT scans are downloaded from the Lung Image Database Consortium (LIDC). Preprocessing is implemented using contrast stretching and enhancing. Lung segmentation and nodule detection stages are performed using a combination of region growing, thresholding and morphological operations. Each 3D structure is then subjected to tabular structure elimination to provide nodule candidates. In the false positive reduction stage, some of the basic nodule features are extracted from the training data to set thresholds for a simple rule-based classifier. The CAD achieved sensitivity of 77.77%, specificity of 69.5% and accuracy 70.53 % with an average 4.1 FPs/scan.

Lung cancer occurs in the lungs, trachea, or bronchi. This cancer is often caused by malignant nodules. These cancer cells spread uncontrollably to other organs of the body and pose a threat to life. An accurate assessment of disease severity is critical to determining the optimal treatment approach. In this study, a Taguchi-based convolutional neural network (CNN) was proposed for classifying nodules into malignant or benign. For setting parameters in a CNN, most users adopt trial and error to determine structural parameters. This study used the Taguchi method for selecting preliminary factors. The orthogonal table design is used in the Taguchi method. The final optimal parameter combination was determined, as were the most significant parameters. To verify the proposed method, the lung image database consortium data set from the National Cancer Institute was used for analysis. The database contains a total of 16,471 images, including 11,139 malignant nodule images. The experimental results demonstrated that the proposed method with the optimal parameter combination obtained an accuracy of 99.6%.

Volumetric lung tumor segmentation and accurate longitudinal tracking of tumor volume changes from computed tomography images are essential for monitoring tumor response to therapy. Hence, we developed two multiple resolution residually connected network (MRRN) formulations called incremental-MRRN and dense-MRRN. Our networks simultaneously combine features across multiple image resolution and feature levels through residual connections to detect and segment the lung tumors. We evaluated our method on a total of 1210 non-small cell (NSCLC) lung tumors and nodules from three data sets consisting of 377 tumors from the open-source Cancer Imaging Archive (TCIA), 304 advanced stage NSCLC treated with anti- PD-1 checkpoint immunotherapy from internal institution MSKCC data set, and 529 lung nodules from the Lung Image Database Consortium (LIDC). The algorithm was trained using 377 tumors from the TCIA data set and validated on the MSKCC and tested on LIDC data sets. The segmentation accuracy compared to expert delineations was evaluated by computing the dice similarity coefficient, Hausdorff distances, sensitivity, and precision metrics. Our best performing incremental-MRRN method produced the highest DSC of 0.74 ± 0.13 for TCIA, 0.75±0.12 for MSKCC, and 0.68±0.23 for the LIDC data sets. There was no significant difference in the estimations of volumetric tumor changes computed using the incremental-MRRN method compared with the expert segmentation. In summary, we have developed a multi-scale CNN approach for volumetrically segmenting lung tumors which enables accurate, automated identification of and serial measurement of tumor volumes in the lung.

We propose a computerized framework that, given a region of interest (ROI) circumscribing a lesion, not only predicts radiological observations related to the lesion characteristics with 83.2% average prediction accuracy but also derives explicit association between low-level imaging features and high-level semantic terms by exploiting their statistical correlation. Such direct association between semantic concepts and low-level imaging features can be leveraged to build a powerful annotation system for radiological images that not only allows the computer to infer the semantics from diverse medical images and run automatic reasoning for making diagnostic decision but also provides “human-interpretable explanation” of the system output to facilitate better end user understanding of computer-based diagnostic decisions. The core component of our framework is a radiological observation detection algorithm that maximizes the low-level imaging feature relevancy for each high-level semantic term. On a liver lesion CT dataset, we have implemented our framework by incorporating a large set of state-of-the-art low-level imaging features. Additionally, we included a novel feature that quantifies lesion(s) present within the liver that have a similar appearance as the primary lesion identified by the radiologist. Our framework achieved a high prediction accuracy (83.2%), and the derived association between semantic concepts and imaging features closely correlates with human expectation. The framework has been only tested on liver lesion CT images, but it is capable of being applied to other imaging domains.

Breast cancer is a leading global cause of cancer-related deaths highlighting the need for an accurate diagnostic system. Up to now, computer-aided diagnosis (CAD) system plays an essential role in supporting pathologists with prompt and accurate classification. Feature selection within the CAD system is crucial as it helps identify the most relevant data for subsequent classification tasks. This paper proposed a novel method that focuses on fuzzy discretization in handling continuous features and selecting relevant and interactive features while eliminating redundancy using Class Association Rule Mining (CARM). The proposed method, FD-CARI was compared with other feature selection techniques including CFS, FCBF, Consistency, Relief-F, and mRMR using five different machine learning classifiers such as Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), Naive Bayes (NB), and Support Vector Machine (SVM). Performance evaluation metrics such as Accuracy (ACC), Sensitivity (SEN), Specificity (SPE), Precision, F1-Score, and AUC were then utilized. Results: The experimental findings consistently showed that the proposed method achieved high performance with an ACC of 96.21%, SEN of 94.26%, and SPE of 97.38% on the SVM classifiers, and an ACC of 96.05%, SEN of 93.82%, and SPE of 97.38% on the LR classifiers. It demonstrated similar effectiveness to Relief-F for DT and RF classifiers. However, FCBF achieved the highest performance on NB with ACC, SEN, and SPE values of 96.21%, 92%, and 96.60%, respectively. The proposed method efficiently selects relevant and interactive features while enabling classifiers to achieve better classification accuracy.

Ontologies represent domain concepts and relations in a form of semantic network. Many research works use ontologies in the information matchmaking and retrieval. This trend is further accelerated by the convergence of various information sources supported by ontologies. In this paper, we propose a novel multi-modality ontology model that integrates both the low-level image features and the high-level text information to represent image contents for image retrieval. By embedding this ontology into an image retrieval system, we are able to realize intelligent image retrieval with high precision. Moreover, benefiting from the soft-coded ontology model, this system has good flexibility and can be easily extended to the larger domains. Currently, our experiment is conducted on the animal domain canine. An ontology has been built based on the low-level features and the domain knowledge of canine. A prototype retrieval system is set up to assess the performance. We compare our experiment results with traditional text-based image search engine and prove the advantages of our approach.

In this paper, we propose a text extraction method from camera-captured document style images and propose a text segmentation method based on a color clustering method. The proposed extraction method detects text regions from the images using two low-level image features and verifies the regions through a high-level text stroke feature. The two level features are combined hierarchically. The low-level features are intensity variation and color variance. And, we use text strokes as a high-level feature using multi-resolution wavelet transforms on local image areas. The stroke feature vector is an input to a SVM (support vector machine) for verification, when needed. The proposed text segmentation method uses color clustering to the extracted text regions. We improved K-means clustering method and it selects K and initial seed values automatically. We tested the proposed methods with various document style images captured by three different cameras. We confirmed that the extraction rates are good enough to be used in real-life applications.

BackgroundEarly diagnosis of pulmonary nodules is essential for effective prevention and treatment of pulmonary cancer. However, the heterogeneous and complex characteristics of pulmonary nodules, such as shape, size, speculation, and texture, present significant challenges in clinical diagnosis, which computer-aided diagnosis (CAD) can help address. Moreover, the varied performance of deep learning methods in CAD and limited model interpretability often hinder clinicians’ understanding of CAD results. In this study, we propose a multitask Swin Transformer (MTST) for classifying benign and malignant pulmonary nodules, which outputs nodule features as classification criteria.MethodsWe introduce a MTST model for feature extraction, designed with a multitask layer that simultaneously outputs benign and malignant binary classification, multilevel classification, and a detailed analysis of pulmonary nodule features. In addition, we incorporate image augmentation using a U-Net generative adversarial network (GAN) model to enhance the training process.ResultsExperimental findings on the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) dataset demonstrate that the proposed MTST outperforms conventional convolutional neural networks (CNN)-based networks across multiple tasks. Specifically, MTST achieved an accuracy of 93.24% in binary classification of benign and malignant nodules and demonstrated superior performance in nodule feature evaluation. For multilevel classification of pulmonary nodules, the Swin Transformer achieved an accuracy of 95.73%. On the training, validation, and test sets (9,600/2,400/1,600 nodules), the MTST model achieved an accuracy of 93.74%, sensitivity of 91.55%, and specificity of 96.09%. The results indicate that the MTST model aligns well with clinical diagnostic practices, offering improved performance and reliability.ConclusionsThe MTST model’s efficacy in binary classification, multiclass classification, and feature evaluation confirms its potential as a valuable tool for CAD systems in clinical settings.

In this paper, we propose a clustering method in order to effectively present image search results on the Web. In order to reflect the difference of image semantics among the images, we use the meta-tags added by social tagging. Furthermore, we use low-level image features and photo metadata in order to consider the image looks and photo-taking conditions. We applied constrained agglomerative clustering method with must-link constraints for better clustering results by using multiple similarity metric. We conducted experiments to demonstrate that the proposed method effectively clusters image search results comparing to the traditional weighted similarity aggregation approach, and in some cases the clustering performance of our approach is better than other existing approaches for clustering tasks on an online photo sharing site.

Shape information plays an important role in biomedical image analysis because of the strong shape characteristics of biological structures. It is often used as a prior to constrain or refine the intermediate shape information derived from low-level image features. In this paper, we give an overview of the sparse shape composition based prior modeling method and its various applications of biomedical image analysis. Instead of learning a generative shape model, it incorporates shape priors on-the-fly through the sparse shape composition. Particularly, a shape instance derived from low level image features is refined by a sparse linear combination of a sparse set of shapes in the repository. We also design three strategies to improve run-time efficiency. 1) When the shape repository contains a large number of instances, K-SVD can be used to learn a more compact but still informative shape dictionary. 2) If the derived shape instance has a large number of vertices, which often appears in 3D problems, affinity propagation method can be used to partition the surface into small subregions. 3) When there are multiple structures, hierarchical scheme is employed to model them simultaneously. These strategies decrease the scale of the sparse optimization problem and thus speed up the algorithm. Our method is applied on different biomedical image analysis problems, including localization, tracking and segmentation of anatomical structures. In all of these applications, this method achieves promising results.

Considering uncertainty within the learning methods is attracting more and more attentions in recent years, yet not much work has been done when it comes to the area of handwritten digit recognition. In this paper, the active belief decision tree is used to handle the epistemic uncertainty in the practical application. The detailed feature extraction method for belief decision trees is proposed and discussed. Under the conditions such as lack of knowledge about true labels, few instances to train the model and low-dimensional image features, we still achieve a good classification performance as shown in experiments with data from the commonly used MNIST handwritten digits benchmark.