XRaySwinGen: Automatic medical reporting for X-ray exams with multimodal model

The importance of radiology in modern medicine is acknowledged for its non-invasive diagnostic capabilities, yet the manual formulation of unstructured medical reports poses time constraints and error risks. This study addresses the common limitation of Artificial Intelligence applications in medical image captioning, which typically focus on classification problems, lacking detailed information about the patient's condition. Despite advancements in AI-generated medical reports that incorporate descriptive details from X-ray images, which are essential for comprehensive reports, the challenge persists. The proposed solution involves a multimodal model utilizing Computer Vision for image representation and Natural Language Processing for textual report generation. A notable contribution is the innovative use of the Swin Transformer as the image encoder, enabling hierarchical mapping and enhanced model perception without a surge in parameters or computational costs. The model incorporates GPT-2 as the textual decoder, integrating cross-attention layers and bilingual training with datasets in Portuguese PT-BR and English. Promising results are noted in the proposed database with ROUGE-L 0.748, METEOR 0.741, and NIH CHEST X-ray with ROUGE-L 0.404 and METEOR 0.393.

A computing platform to analyze breast abnormalities using infrared images

The present work shows a computational tool developed in the MATLAB platform. Its main functionality is to evaluate a thermal model of the breast. This computational infrastructure consists of modules in which manipulate the infrared images and calculate breast temperature profiles. It also allows the analysis of breast nodules. The different modules of the framework are interconnected through an interface which the major purpose is to automatize the whole process of the infrared image analysis, in a quick and organized way. The tool is initially supplied with a three-dimensional mesh that represents the substitute geometry of the patient's breast together with her infrared images which are transformed into temperature matrices. Through these matrices, the frontal and lateral mappings are performed by specified modules. This process generates an image and a text file with all the temperatures associated to the nodes of the surface mesh. The developed tool is also able to manage the use of a commercial mesh generation program and a computational fluid dynamics code, the FLUENT, in order to validate the technique by the use of a parametric analysis. In these analyses, the tumor may have several geometric shapes and different locations within the breast.

Computational method for aid in the diagnosis of sixth optic nerve palsy through digital videos

The sixth cranial nerve, also known as the abducens nerve, is responsible for controlling the movements of the lateral rectus muscle. Palsies on the sixth nerve prevent some muscles that control eye movements from proper functioning, causing headaches, migraines, blurred vision, vertigo, and double vision. Hence, such palsy should be diagnosed in the early stages to treat it without leaving any sequela. The usual methods for diagnosing the sixth nerve palsy are invasive or depend on expensive equipment, and computer-based methods designed specifically to diagnose the aforementioned palsy were not found until the publication of this work. Therefore, a low-cost, non-invasive method can support or guide the ophthalmologist's diagnosis. In this context, this work presents a computational methodology to aid in diagnosing the sixth nerve palsy using videos to assist ophthalmologists in the diagnostic process, serving as a second opinion. The proposed method uses convolutional neural networks and image processing techniques to track both eyes' movement trajectory during the video. With this trajectory, it is possible to calculate the average velocity (AV) in which each eye moves. Since it is known that paretic eyes move slower than healthy eyes, comparing the AV of both eyes can determine if the eye is healthy or paretic. The results obtained with the proposed method showed that paretic eyes move at least 19.65% slower than healthy ones. This threshold, along with the AV of the movement of the eyes, can help ophthalmologists in their analysis. The proposed method reached 92.64% accuracy in diagnosing the sixth optic nerve palsy (SONP), with a Kappa index of 0.925, which highlights the reliability of the results and gives favorable perspectives for further clinical application.

Liver segmentation from computed tomography images using cascade deep learning

BACKGROUND

Liver segmentation is a fundamental step in the treatment planning and diagnosis of liver cancer. However, manual segmentation of liver is time-consuming because of the large slice quantity and subjectiveness associated with the specialist's experience, which can lead to segmentation errors. Thus, the segmentation process can be automated using computational methods for better time efficiency and accuracy. However, automatic liver segmentation is a challenging task, as the liver can vary in shape, ill-defined borders, and lesions, which affect its appearance. We aim to propose an automatic method for liver segmentation using computed tomography (CT) images.

METHODS

The proposed method, based on deep convolutional neural network models and image processing techniques, comprise of four main steps: (1) image preprocessing, (2) initial segmentation, (3) reconstruction, and (4) final segmentation.

RESULTS

We evaluated the proposed method using 131 CT images from the LiTS image base. An average sensitivity of 95.45%, an average specificity of 99.86%, an average Dice coefficient of 95.64%, an average volumetric overlap error (VOE) of 8.28%, an average relative volume difference (RVD) of -0.41%, and an average Hausdorff distance (HD) of 26.60 mm were achieved.

CONCLUSIONS

This study demonstrates that liver segmentation, even when lesions are present in CT images, can be efficiently performed using a cascade approach and including a reconstruction step based on deep convolutional neural networks.

Automatic method for classifying COVID-19 patients based on chest X-ray images, using deep features and PSO-optimized XGBoost

The COVID-19 pandemic, which originated in December 2019 in the city of Wuhan, China, continues to have a devastating effect on the health and well-being of the global population. Currently, approximately 8.8 million people have already been infected and more than 465,740 people have died worldwide. An important step in combating COVID-19 is the screening of infected patients using chest X-ray (CXR) images. However, this task is extremely time-consuming and prone to variability among specialists owing to its heterogeneity. Therefore, the present study aims to assist specialists in identifying COVID-19 patients from their chest radiographs, using automated computational techniques. The proposed method has four main steps: (1) the acquisition of the dataset, from two public databases; (2) the standardization of images through preprocessing; (3) the extraction of features using a deep features-based approach implemented through the networks VGG19, Inception-v3, and ResNet50; (4) the classifying of images into COVID-19 groups, using eXtreme Gradient Boosting (XGBoost) optimized by particle swarm optimization (PSO). In the best-case scenario, the proposed method achieved an accuracy of 98.71%, a precision of 98.89%, a recall of 99.63%, and an F1-score of 99.25%. In our study, we demonstrated that the problem of classifying CXR images of patients under COVID-19 and non-COVID-19 conditions can be solved efficiently by combining a deep features-based approach with a robust classifier (XGBoost) optimized by an evolutionary algorithm (PSO). The proposed method offers considerable advantages for clinicians seeking to tackle the current COVID-19 pandemic.

Bayesian convolutional neural network estimation for pediatric pneumonia detection and diagnosis

BACKGROUND AND OBJECTIVES

Pneumonia is a disease that affects the lungs, making breathing difficult. Nowadays, pneumonia is the disease that kills the most children under the age of five in the world, and if no action is taken, pneumonia is estimated to kill 11 million children by the year 2030. Knowing that rapid and accurate diagnosis of pneumonia is a significant factor in reducing mortality, acceleration, or automation of the diagnostic process is highly desirable. The use of computational methods can decrease specialists' workload and even offer a second opinion, increasing the number of accurate diagnostics.

METHODS

This work proposes a method for constructing a specific convolutional neural network architecture to detect pneumonia and classify viral and bacterial types using Bayesian optimization from pre-trained networks.

RESULTS

The results obtained are promising, in the order of 0.964 accuracy for pneumonia detection and 0.957 accuracy for pneumonia type classification.

CONCLUSION

This research demonstrated the efficiency of CNN architecture estimation for detecting and diagnosing pneumonia using Bayesian optimization. The proposed network proved to have promising results, despite not using common preprocessing techniques such as histogram equalization and lung segmentation. This fact shows that the proposed method provides efficient and high-performance neural networks since image preprocessing is unnecessary.

Surgical planning of horizontal strabismus using multiple output regression tree

Strabismus is an eye disease that affects about 0.12%-9.86% of the population, which can cause irreversible sensory damage to vision and psychological problems. The most severe cases require surgical intervention, despite other less invasive techniques being available for a more conservative approach. As for surgeries, the treatment goal is to align the eyes to recover binocular vision, which demands knowledge, training, and experience. One of the leading causes of failure is human error during the measurement of deviation. Thus, this work proposes a new method based on the Decision Tree Regressor algorithms to assist in the surgical planning for horizontal strabismus to predict recoil and resection measures in the lateral and medial rectus muscles. In the presented method, two application approaches were taken, being in the form of multiple single target models, one procedure at a time, and the form of one multiple target model or all surgical procedures together. The method's efficiency is indicated by the average difference between the value indicated by the method and the physician's value. In our most accurate model, an average error of 0.66 mm was obtained for all surgical procedures, both for resection and recoil in the indication of the horizontal strabismus surgical planning. The results present the feasibility of using Decision Tree Regressor algorithms to perform the planning of strabismus surgeries, making it possible to predict correction values for surgical procedures based on medical data analysis and exceeding state-of-art.

Tear Film Classification in Interferometry Eye Images Using Phylogenetic Diversity Indexes and Ripley's K Function

Dry eye syndrome is one of the most frequently reported eye diseases in ophthalmological practice. The diagnosis of this disease is a challenging task due to its multifactorial etiology. One of the most applied tests is the manual classification of tear film images captured with the Doane interferometer. The interference phenomena in these images can be characterized as texture patterns, which can be automatically classified into one of the following categories: strong fringes, coalescing strong fringes, fine fringes, coalescing fine fringes, and debris. This work presents a method for classifying tear film images based on texture analysis using phylogenetic diversity indexes and Ripley's K function. The proposed method consists of six main steps: acquisition of the image dataset; segmentation of the region of interest; feature extraction using phylogenetic diversity indexes and Ripley's K function; feature selection using Greedy Stepwise; classification using the algorithms Support Vector Machine (SVM), Random Forest (RF), Naive Bayes (NB), Multilayer Perceptron (MLP), Random Tree (RT) and Radial Basis Function Network (RBFNet); and (6) validation of results. The best result, using the RF classifier, we obtained classification rates higher than 99% of accuracy with 0.843% of standard deviation, 0.999 of the area under the Receiver Operating Characteristics (ROC) curve, 0.995 of Kappa and 0.996 of F-Measure. The experimental results demonstrate that the proposed method is promising and can potentially be used by experts to accurately diagnose dry eye syndrome in tear film images.

Esophagus segmentation from planning CT images using an atlas-based deep learning approach

BACKGROUND AND OBJECTIVE

One of the main steps in the planning of radiotherapy (RT) is the segmentation of organs at risk (OARs) in Computed Tomography (CT). The esophagus is one of the most difficult OARs to segment. The boundaries between the esophagus and other surrounding tissues are not well-defined, and it is presented in several slices of the CT. Thus, manually segment the esophagus requires a lot of experience and takes time. This difficulty in manual segmentation combined with fatigue due to the number of slices to segment can cause human errors. To address these challenges, computational solutions for analyzing medical images and proposing automated segmentation have been developed and explored in recent years. In this work, we propose a fully automatic method for esophagus segmentation for better planning of radiotherapy in CT.

METHODS

The proposed method is a fully automated segmentation of the esophagus, consisting of 5 main steps: (a) image acquisition; (b) VOI segmentation; (c) preprocessing; (d) esophagus segmentation; and (e) segmentation refinement.

RESULTS

The method was applied in a database of 36 CT acquired from 3 different institutes. It achieved the best results in literature so far: Dice coefficient value of 82.15%, Jaccard Index of 70.21%, accuracy of 99.69%, sensitivity of 90.61%, specificity of 99.76%, and Hausdorff Distance of 6.1030 mm.

CONCLUSIONS

With the achieved results, we were able to show how promising the method is, and that applying it in large medical centers, where esophagus segmentation is still an arduous and challenging task, can be of great help to the specialists.

Interferometer eye image classification for dry eye categorization using phylogenetic diversity indexes for texture analysis

Background and Objective Dry eye syndrome disease negatively impacts many people in various ways. Several tests are required to diagnose it for evaluating different physiological characteristics. One of the most applied tests for this is the manual classification of tear film images captured with Doane interferometer. Interferometry images can be categorized into five groups: debris, fine fringes, coalescing fine fringes, strong fringes, and coalescing strong fringes. Instability in the tear film creates the need for an automatic system to provide experts with diagnostic support. Therefore, the purpose of this study was to propose a method for automatic classification of the tear film lipid layer using phylogenetic diversity indexes for feature extraction and several classifiers. Methods The proposed method consisted of five main steps: (1) acquisition of VOPTICAL_GCU image dataset, (2) segmentation of the region of interest, (3) feature extraction using phylogenetic diversity indexes, (4) classification using the algorithms Support Vector Machines, Random Forest, Naive Bayes, Multilayer Perceptron, Random Tree, and RBFNetwork, and, (5) validation of results. Results The best result was obtained using Random Forest classifier, reaching an accuracy of over 97%, standard deviation of 0.51%, an area under the receiver operating characteristic curve of 0.99, a Kappa index of 0.96, and an F-Measure of 0.97. Conclusions The proposed method demonstrated that the tear film lipid layer classification problem can be resolved efficiently by using phylogenetic diversity indexes.

An automatic method for lung segmentation and reconstruction in chest X-ray using deep neural networks

BACKGROUND AND OBJECTIVE

Chest X-ray (CXR) is one of the most used imaging techniques for detection and diagnosis of pulmonary diseases. A critical component in any computer-aided system, for either detection or diagnosis in digital CXR, is the automatic segmentation of the lung field. One of the main challenges inherent to this task is to include in the segmentation the lung regions overlapped by dense abnormalities, also known as opacities, which can be caused by diseases such as tuberculosis and pneumonia. This specific task is difficult because opacities frequently reach high intensity values which can be incorrectly interpreted by an automatic method as the lung boundary, and as a consequence, this creates a challenge in the segmentation process, because the chances of incomplete segmentations are increased considerably. The purpose of this work is to propose a method for automatic segmentation of lungs in CXR that addresses this problem by reconstructing the lung regions "lost" due to pulmonary abnormalities.

METHODS

The proposed method, which features two deep convolutional neural network models, consists of four steps main steps: (1) image acquisition, (2) initial segmentation, (3) reconstruction and (4) final segmentation.

RESULTS

The proposed method was experimented on 138 Chest X-ray images from Montgomery County's Tuberculosis Control Program, and has achieved as best result an average sensitivity of 97.54%, an average specificity of 96.79%, an average accuracy of 96.97%, an average Dice coefficient of 94%, and an average Jaccard index of 88.07%.

CONCLUSIONS

We demonstrate in our lung segmentation method that the problem of dense abnormalities in Chest X-rays can be efficiently addressed by performing a reconstruction step based on a deep convolutional neural network model.

Modified Quality Threshold Clustering for Temporal Analysis and Classification of Lung Lesions

Lung cancer is the type of cancer that most often kills after the initial diagnosis. To aid the specialist in its diagnosis, temporal evaluation is a potential tool for analyzing indeterminate lesions, which may be benign or malignant, during treatment. With this goal in mind, a methodology is herein proposed for the analysis, quantification, and visualization of changes in lung lesions. This methodology uses a modified version of the quality threshold clustering algorithm to associate each voxel of the lesion to a cluster, and changes in the lesion over time are defined in terms of voxel moves to another cluster. In addition, statistical features are extracted for classification of the lesion as benign or malignant. To develop the proposed methodology, two databases of pulmonary lesions were used, one for malignant lesions in treatment (public) and the other for indeterminate cases (private). We determined that the density change percentage varied from 6.22% to 36.93% of lesion volume in the public database of malignant lesions under treatment and from 19.98% to 38.81% in the private database of lung nodules. Additionally, other inter-cluster density change measures were obtained. These measures indicate the degree of change in the clusters and how each of them is abundant in relation to volume. From the statistical analysis of regions in which the density changes occurred, we were able to discriminate lung lesions with an accuracy of 98.41%, demonstrating that these changes can indicate the true nature of the lesion. In addition to visualizing the density changes occurring in lesions over time, we quantified these changes and analyzed the entire set through volumetry, which is the technique most commonly used to analyze changes in pulmonary lesions.

Convolutional neural network-based PSO for lung nodule false positive reduction on CT images

BACKGROUND AND OBJECTIVE

Detection of lung nodules is critical in CAD systems; this is because of their similar contrast with other structures and low density, which result in the generation of numerous false positives (FPs). Therefore, this study proposes a methodology to reduce the FP number using a deep learning technique in conjunction with an evolutionary technique.

METHOD

The particle swarm optimization (PSO) algorithm was used to optimize the network hyperparameters in the convolutional neural network (CNN) in order to enhance the network performance and eliminate the requirement of manual search.

RESULTS

The methodology was tested on computed tomography (CT) scans from the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) with the highest accuracy of 97.62%, sensitivity of 92.20%, specificity of 98.64%, and area under the receiver operating characteristic (ROC) curve of 0.955.

CONCLUSION

The results demonstrate the high performance-potential of the PSO algorithm in the identification of optimal CNN hyperparameters for lung nodule candidate classification into nodules and non-nodules, increasing the sensitivity rates in the FP reduction step of CAD systems.

Detection of mass regions in mammograms by bilateral analysis adapted to breast density using similarity indexes and convolutional neural networks

BACKGROUND AND OBJECTIVE

The processing of medical image is an important tool to assist in minimizing the degree of uncertainty of the specialist, while providing specialists with an additional source of detect and diagnosis information. Breast cancer is the most common type of cancer that affects the female population around the world. It is also the most deadly type of cancer among women. It is the second most common type of cancer among all others. The most common examination to diagnose breast cancer early is mammography. In the last decades, computational techniques have been developed with the purpose of automatically detecting structures that maybe associated with tumors in mammography examination. This work presents a computational methodology to automatically detection of mass regions in mammography by using a convolutional neural network.

METHODS

The materials used in this work is the DDSM database. The method proposed consists of two phases: training phase and test phase. The training phase has 2 main steps: (1) create a model to classify breast tissue into dense and non-dense (2) create a model to classify regions of breast into mass and non-mass. The test phase has 7 step: (1) preprocessing; (2) registration; (3) segmentation; (4) first reduction of false positives; (5) preprocessing of regions segmented; (6) density tissue classification (7) second reduction of false positives where regions will be classified into mass and non-mass.

RESULTS

The proposed method achieved 95.6% of accuracy in classify non-dense breasts tissue and 97,72% accuracy in classify dense breasts. To detect regions of mass in non-dense breast, the method achieved a sensitivity value of 91.5%, and specificity value of 90.7%, with 91% accuracy. To detect regions in dense breasts, our method achieved 90.4% of sensitivity and 96.4% of specificity, with accuracy of 94.8%.

CONCLUSIONS

According to the results achieved by CNN, we demonstrate the feasibility of using convolutional neural networks on medical image processing techniques for classification of breast tissue and mass detection.

Computer-Aided Diagnosis of Lung Nodules in Computed Tomography by Using Phylogenetic Diversity, Genetic Algorithm, and SVM

Lung cancer is pointed as the major cause of death among patients with cancer throughout the world. This work is intended to develop a methodology for diagnosis of lung nodules using images from the Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI). The proposed methodology uses image processing and pattern recognition techniques. In order to differentiate between the patterns of malignant and benign nodules, we used phylogenetic diversity by means of particular indexes, that are: intensive quadratic entropy, extensive quadratic entropy, average taxonomic distinctness, total taxonomic distinctness, and pure diversity indexes. After that, we applied the genetic algorithm for selection of the best model. In the tests' stage, we applied the proposed methodology to 1405 (394 malignant and 1011 benign) nodules. The proposed work presents promising results at the classification into malignant and benign, achieving accuracy of 92.52%, sensitivity of 93.1% and specificity of 92.26%. The results demonstrated a good rate of correct detections using texture features. Since a precocious detection allows a faster therapeutic intervention, thus a more favorable prognostic to the patient, we propose herein a methodology that contributes to the area in this aspect.

Statistical tools for the temporal analysis and classification of lung lesions

BACKGROUND AND OBJECTIVE

Lung cancer remains one of the most common cancers globally. Temporal evaluation is an important tool for analyzing the malignant behavior of lesions during treatment, or of indeterminate lesions that may be benign. This work proposes a methodology for the analysis, quantification, and visualization of small (local) and large (global) changes in lung lesions. In addition, we extract textural features for the classification of lesions as benign or malignant.

METHODS

We employ the statistical concept of uncertainty to associate each voxel of a lesion to a probability that changes occur in the lesion over time. We employ the Jensen divergence and hypothesis test locally to verify voxel-to-voxel changes, and globally to capture changes in lesion volumes.

RESULTS

For the local hypothesis test, we determine that the change in density varies by between 3.84 and 40.01% of the lesion volume in a public database of malignant lesions under treatment, and by between 5.76 and 35.43% in a private database of benign lung nodules. From the texture analysis of regions in which the density changes occur, we are able to discriminate lung lesions with an accuracy of 98.41%, which shows that these changes can indicate the true nature of the lesion.

CONCLUSION

In addition to the visual aspects of the density changes occurring in the lesions over time, we quantify these changes and analyze the entire set using volumetry. In the case of malignant lesions, large b-divergence values are associated with major changes in lesion volume. In addition, this occurs when the change in volume is small but is associated with significant changes in density, as indicated by the histogram divergence. For benign lesions, the methodology shows that even in cases where the change in volume is small, a change of density occurs. This proves that even in lesions that appear stable, a change in density occurs.

Automatic Detection of Masses in Mammograms Using Quality Threshold Clustering, Correlogram Function, and SVM

Breast cancer is the second most common type of cancer in the world. Several computer-aided detection and diagnosis systems have been used to assist health experts and to indicate suspect areas that would be difficult to perceive by the human eye; this approach has aided in the detection and diagnosis of cancer. The present work proposes a method for the automatic detection of masses in digital mammograms by using quality threshold (QT), a correlogram function, and the support vector machine (SVM). This methodology comprises the following steps: The first step is to perform preprocessing with a low-pass filter, which increases the scale of the contrast, and the next step is to use an enhancement to the wavelet transform with a linear function. After the preprocessing is segmentation using QT; then, we perform post-processing, which involves the selection of the best mass candidates. This step is performed by analyzing the shape descriptors through the SVM. For the stage that involves the extraction of texture features, we used Haralick descriptors and a correlogram function. In the classification stage, the SVM was again used for training, validation, and final test. The results were as follows: sensitivity 92.31 %, specificity 82.2 %, accuracy 83.53 %, mean rate of false positives per image 1.12, and area under the receiver operating characteristic (ROC) curve 0.8033. Breast cancer is notable for presenting the highest mortality rate in addition to one of the smallest survival rates after diagnosis. An early diagnosis means a considerable increase in the survival chance of the patients. The methodology proposed herein contributes to the early diagnosis and survival rate and, thus, proves to be a useful tool for specialists who attempt to anticipate the detection of masses.

Computer-Aided Detection and Diagnosis of Breast Cancer Using Machine Learning, Texture and Shape Features

Breast cancer is a malignant (cancer) tumor that starts from cells of the breast, being the major cause of deaths by cancer in the female population. There has been tremendous interest in the use of image processing and analysis techniques for computer aided detection (CAD)/ diagnostics (CADx) in digital mammograms. The goal has been to increase diagnostic accuracy as well as the reproducibility of mammographic interpretation. CAD/CADx systems can aid radiologists by providing a second opinion and may be used in the first stage of examination in the near future, providing the reduction of the variability among radiologists in the interpretation of mammograms. This chapter provides an overview of techniques used in computer-aided detection and diagnosis of breast cancer. The authors focus on the application of texture and shape tissues signature used with machine learning techniques, like support vector machines (SVM) and growing neural gas (GNG).

Computational methodology for automatic detection of strabismus in digital images through Hirschberg test

Strabismus is a pathology that affects about 4% of the population, causing aesthetic problems, reversible at any age; however, problems that can also cause irreversible muscular alterations, and alter the vision mechanism. The Hirschberg test is one of the exams used to detect this pathology. The application of high technology resources to help diagnose and treat ophthalmological conditions is, lamentably, not commonly found in the sub-specialty of strabismus. This work presents a methodology for automatic detection of strabismus in digital images through the Hirschberg test. For such, the work was organized into four stages: (1) finding the region of the eyes; (2) determining the precise location of the eyes; (3) locating the limbus and brightness; and (4) identifying strabismus. The methodology has produced results on the range of 100% sensibility, 91.3% specificity and 94% for the correct identification of strabismus, ensuring the efficiency of its geostatistical functions for the extraction of eye texture and for the calculation of the alignment between the eyes on digital images obtained from the Hirschberg test.

Methodology for automatic detection of lung nodules in computerized tomography images

Lung cancer is a disease with significant prevalence in several countries around the world. Its difficult treatment and rapid progression make the mortality rates among people affected by this illness to be very high. Aiming to offer a computational alternative for helping in detection of nodules, serving as a second opinion to the specialists, this work proposes a totally automatic methodology based on successive detection refining stages. The automated lung nodules detection scheme consists of six stages: thorax extraction, lung extraction, lung reconstruction, structures extraction, tubular structures elimination, and false positive reduction. In the thorax extraction stage all the artifacts external to the patient's body are discarded. Lung extraction stage is responsible for the identification of the lung parenchyma. The objective of the lung reconstruction stage is to prevent incorrect elimination of portions belonging to the parenchyma. Structures extraction stage comprises the selection of dense structures from inside the lung parenchyma. The next stage, tubular structures elimination eliminates a great part of the pulmonary trees. Finally, the false positive stage selects only structures with great probability to be nodule. Each of the several stages has very specific objectives in detection of particular cases of lung nodules, ensuring good matching rates even in difficult detection situations. We use 33 exams with diversified diagnosis and slices numbers for validating the methodology. We obtained a false positive per exam rate of 0.42 and false negative rate of 0.15. The total classification sensitivity obtained, measured out of the nodule candidates, was 84.84%. The specificity achieved was 96.15% and the total accuracy of the method was 95.21%.

Diagnosis of solitary lung nodules using the local form of Ripley's K function applied to three-dimensional CT data

This paper analyzes the application of Ripley's K function to characterize lung nodules as malignant or benign in computerized tomography images. The proposed characterization method is based on a selection of measures from Ripley's K function to discriminate between benign and malignant nodules, using stepwise discriminant analysis. Based on the selected measures, a linear discriminant analysis procedure is performed once again in order to predict the classification of each nodule. To evaluate the ability of these features to discriminate the nodules, a set of tests was carried out using a sample of 39 pulmonary nodules, 29 benign and 10 malignant. A leave-one-out procedure was used to provide a less biased estimate of the linear discriminator's performance. The best setting of the analyzed function in the tested sample presented 70.0% of sensitivity but with 100.0% of specificity and 92.3% of accuracy. Thus, preliminary results of this approach are very promising regarding its contribution to the diagnosis of pulmonary nodules, but it still needs to be tested with larger series and associated to other quantitative imaging methods in order to improve global performance.