Computer-aided Detection Method Research Articles

Pulmonary nodule detection in low dose computed tomography using a medical-to-medical transfer learning approach.

Lung cancer is the second most common cancer and the leading cause of cancer death globally. Low dose computed tomography (LDCT) is the recommended imaging screening tool for the early detection of lung cancer. A fully automated computer-aided detection method for LDCT will greatly improve the existing clinical workflow. Most of the existing methods for lung detection are designed for high-dose CTs (HDCTs), and those methods cannot be directly applied to LDCTs due to domain shifts and inferior quality of LDCT images. In this work, we describe a semi-automated transfer learning-based approach for the early detection of lung nodules using LDCTs. In this work, we developed an algorithm based on the object detection model, you only look once (YOLO) to detect lung nodules. The YOLO model was first trained on CTs, and the pre-trained weights were used as initial weights during the retraining of the model on LDCTs using a medical-to-medical transfer learning approach. The dataset for this study was from a screening trial consisting of LDCTs acquired from 50 biopsy-confirmed lung cancer patients obtained over 3 consecutive years (T1, T2, and T3). About 60 lung cancer patients' HDCTs were obtained from a public dataset. The developed model was evaluated using a hold-out test set comprising 15 patient cases (93 slices with cancerous nodules) using precision, specificity, recall, and F1-score. The evaluation metrics were reported patient-wise on a per-year basis and averaged for 3 years. For comparative analysis, the proposed detection model was trained using pre-trained weights from the COCO dataset as the initial weights. A paired t-test and chi-squared test with an alpha value of 0.05 were used for statistical significance testing. The results were reported by comparing the proposed model developed using HDCT pre-trained weights with COCO pre-trained weights. The former approach versus the latter approach obtained a precision of 0.982 versus 0.93 in detecting cancerous nodules, specificity of 0.923 versus 0.849 in identifying slices with no cancerous nodules, recall of 0.87 versus 0.886, and F1-score of 0.924 versus 0.903. As the nodule progressed, the former approach achieved a precision of 1, specificity of 0.92, and sensitivity of 0.930. The statistical analysis performed in the comparative study resulted in a -value of 0.0054 for precision and a -value of 0.00034 for specificity. In this study, a semi-automated method was developed to detect lung nodules in LDCTs using HDCT pre-trained weights as the initial weights and retraining the model. Further, the results were compared by replacing HDCT pre-trained weights in the above approach with COCO pre-trained weights. The proposed method may identify early lung nodules during the screening program, reduce overdiagnosis and follow-ups due to misdiagnosis in LDCTs, start treatment options in the affected patients, and lower the mortality rate.

Pulmonary nodule detection in low dose computed tomography using a medical-to-medical transfer learning approach.

Lung cancer is the second most common cancer and the leading cause of cancer death globally. Low dose computed tomography (LDCT) is the recommended imaging screening tool for the early detection of lung cancer. A fully automated computer-aided detection method for LDCT will greatly improve the existing clinical workflow. Most of the existing methods for lung detection are designed for high-dose CTs (HDCTs), and those methods cannot be directly applied to LDCTs due to domain shifts and inferior quality of LDCT images. In this work, we describe a semi-automated transfer learning-based approach for the early detection of lung nodules using LDCTs. In this work, we developed an algorithm based on the object detection model, you only look once (YOLO) to detect lung nodules. The YOLO model was first trained on CTs, and the pre-trained weights were used as initial weights during the retraining of the model on LDCTs using a medical-to-medical transfer learning approach. The dataset for this study was from a screening trial consisting of LDCTs acquired from 50 biopsy-confirmed lung cancer patients obtained over 3 consecutive years (T1, T2, and T3). About 60 lung cancer patients' HDCTs were obtained from a public dataset. The developed model was evaluated using a hold-out test set comprising 15 patient cases (93 slices with cancerous nodules) using precision, specificity, recall, and F1-score. The evaluation metrics were reported patient-wise on a per-year basis and averaged for 3 years. For comparative analysis, the proposed detection model was trained using pre-trained weights from the COCO dataset as the initial weights. A paired t-test and chi-squared test with an alpha value of 0.05 were used for statistical significance testing. The results were reported by comparing the proposed model developed using HDCT pre-trained weights with COCO pre-trained weights. The former approach versus the latter approach obtained a precision of 0.982 versus 0.93 in detecting cancerous nodules, specificity of 0.923 versus 0.849 in identifying slices with no cancerous nodules, recall of 0.87 versus 0.886, and F1-score of 0.924 versus 0.903. As the nodule progressed, the former approach achieved a precision of 1, specificity of 0.92, and sensitivity of 0.930. The statistical analysis performed in the comparative study resulted in a -value of 0.0054 for precision and a -value of 0.00034 for specificity. In this study, a semi-automated method was developed to detect lung nodules in LDCTs using HDCT pre-trained weights as the initial weights and retraining the model. Further, the results were compared by replacing HDCT pre-trained weights in the above approach with COCO pre-trained weights. The proposed method may identify early lung nodules during the screening program, reduce overdiagnosis and follow-ups due to misdiagnosis in LDCTs, start treatment options in the affected patients, and lower the mortality rate.

A 2D-CNN-Based Two-Stage Structural Damage Localization and Quantification Technique Using Time Domain Vibration Data

The conventional approaches for detecting structural degradation are time-consuming, labor-intensive, and costly. The physical monitoring of the structure also poses risks to the health and safety of supervisors. Therefore, damage estimation of any structure using artificial intelligence (AI), more specifically deep learning (DL), is becoming more significant in civil infrastructure. In the presented research article, an efficient two-stage damage detection method is proposed for structural damage detection (SDD) from time domain vibration signals. The proposed method utilizes two-dimensional convolutional neural network (2D-CNN) architecture as a DL algorithm for damage detection. Here, a computer-aided damage detection method for steel beam and frame-type structures is developed using 2D-CNN algorithm in the Google Colab platform. The effectiveness of the proposed method is first verified, and it provides more than 90% accuracy for identifying the damage location and severity of a cantilever beam for single- and multi-damage scenarios from numerically simulated noisy displacement data. The algorithm is also experimentally validated through the raw acceleration data of damaged steel frame joints collected from the Qatar University Grandstand simulator (QUGS). The proposed 2D-CNN algorithm performs better than other DL algorithms by achieving 100% accuracy within 10 epochs for damage detection of steel frames using QUGS data. It demonstrates significant potential for detecting damage location and quantifying damages for single- and multi-damage scenarios using noise-free and noisy datasets. The primary contribution of this study resides in the implementation of two-stage damage detection algorithm utilizing 2D-CNN with time domain vibration response for multiclass damage identification and quantification.

Computer-Aided Detection of Bone Metastasis using Bone Scintigraphy Scans

Abstract: When cancer cells spread from their original site to the bone then it is known as ‘Bone Metastasis’. Patients with solid tumors such as lung, breast, kidney, prostate etc. oftentimes develop bone metastasis. Cancer cells that spread to the bone mostly affect limbs, pelvis, spine, skull and ribs. Once tumors have metastasized to the bone, patients usually undergo skeletal complications collectively known as skeletal-related events (SRE) like skeletal remodelling, fractures, spinal cord compression, pain or anaemia and bowel incontinence. Therefore, early diagnosis of bone metastasis is crucial for increasing survivability as it has a major impact on treatment decision and at the same time it is a decisive factor in the evolution of the disease and the quality of life. ‘Bone Scintigraphy’ is accepted to be an efficient diagnostic tool for whole-body examination of bone metastasis. Computer-Aided Detection method is used to avoid mistakes which might be caused by human interference & experimental sources. CAD system provides faster and more precise results that facilitates in the early detection of bone metastasis, hence increasing the patient’s chance of survival. To reduce the dependency of the system on source of the input image & for making the system robust we have explored various done image enhancement techniques.

Automated unruptured cerebral aneurysms detection in TOF MR angiography images using dual-channel SE-3D UNet: a multi-center research.

Time of flight magnetic resonance angiography (TOF-MRA) is the primary non-invasive screening method for cerebral aneurysms. We aimed to develop a computer-aided aneurysm detection method to improve the diagnostic efficiency and accuracy, especially decrease the false positive rate. This is a retrospective multicenter study. The dataset contained 1160 TOF-MRA examinations composed of unruptured aneurysms (n = 1096) and normal controls (n = 166) from six hospitals. A total of 1037 examinations acquired from 2013 to 2019 were used as training set; 123 examinations acquired from 2020 to 2021 were used as external test set. We proposed an equalized augmentation strategy based on aneurysm location and constructed a detection model based on dual channel SE-3D UNet. The model was trained with a 5-fold cross-validation in the training set, then tested on the external test set. The proposed method achieved 82.46% sensitivity on patient-level, 73.85% sensitivity on lesion-level, and 0.88 false positives per case in the external test set. The performance did not show significant differences in subgroups according to the aneurysm site (except ACA), aneurysm size (except smaller than 3 mm), or MRI scanners. The performance preceded the basic SE-3D UNet by increasing 15.79% patient-level sensitivity and decreasing 4.19 FPs/case. The proposed automated aneurysm detection method achieved acceptable sensitivity while controlling fairly low false positives per case. It might provide a useful auxiliary tool of cerebral aneurysms MRA screening. • The need for automated cerebral aneurysms detecting is growing. • The strategy of equalized augmentation based on aneurysm location and dual-channel input could improve the model performance. • The retrospective multi-center study showed that the proposed automated cerebral aneurysms detection using dual-channel SE-3D UNet could achieve acceptable sensitivity while controlling a low false positive rate.

DSA Image Analysis of Clinical Features and Nursing Care of Cerebral Aneurysm Patients Based on the Deep Learning Algorithm.

Objective A deep learning algorithm was developed for automatic detection and localization of intracranial aneurysms in DSA, and its clinical characteristics were analyzed, and targeted nursing measures were formulated. Methods Using a retrospective multicenter study method based on radiology reports, DSA images of aneurysms were randomly divided into 75 cases in the training set, 20 cases in the internal test set, and 35 cases in the external test set. Using a computer-aided detection method based on the three-dimensional U-Net (3D U-Net), after preprocessing DSA images, automatic segmentation of intracranial blood vessels is performed to obtain regions of interest, and based on the segmentation results, physicians' annotations are introduced. The 3D U-Net network model is trained and adjusted, and the obtained model is used to automatically detect the cerebral aneurysm area. Results Fivefold cross-validation was used for the training set and the internal test set, and a sensitivity of (94.4 ± 1.1)% was obtained. Automatic detection of aneurysms was performed on the external test set, and the average false positive rate was 0.86 FPs/case (false positives/case). The resulting sensitivity was 82.9%. The classification comparison of external test sets showed that the sensitivity of the method for detecting aneurysms with sizes of 5.00~<10.00 mm and ≥10.00 mm (88.2% and 100.0%) was higher than that for aneurysms with sizes of <3.00 mm and 3.00~<5.00 mm (50.0% and 72.7%). The sensitivity of patients aged 50-60 years and >60 years (90.0% and 87.5%) was higher than that of patients aged <50 years (66.7%), and there was little difference between different genders (84.6% in males and 81.8% in females). Conclusion The deep learning algorithm has high diagnostic performance in detecting intracranial aneurysms, which is verified by external datasets.

Automatic esophagus Z‐line delineation in endoscopic images using a new boundary linking method

Due to the American cancer society, many people with esophageal adenocarcinoma are not survived. The treatment rate can be significant in the early detection of Barrett's esophagus (BE) as a premalignant stage for adenocarcinoma. An important landmark to detect BE is the Z-line. BE segmentation is already highly dependent upon the operator's knowledge and skill. The main aim of this study is automatic Z-line extraction using endoscopic images leading to segmentation of the early BE stage. To this end, a computer-aided detection method exploiting k-means clustering, image segmentation using the edge detector, and a novel boundary linking algorithm is proposed. For the evaluation, the gold standard is considered the average contours of Z-lines extracted by the three experts. The proposed method annotated the Z-line with the accuracy and precision of 0.92 and 0.87, respectively, and the value of the average boundary distance is 5.9 pixels. To the results and visual inspection, the presented method can be used for efficient and robust extraction of the Z-line at the early BE stage. Furthermore, it can be used in other medical imaging applications with complex boundaries and low contrast in the images, limiting the common automatic boundary detection methods.

An interpretable deep learning workflow for discovering subvisual abnormalities in CT scans of COVID-19 inpatients and survivors

Tremendous efforts have been made to improve diagnosis and treatment of COVID-19, but knowledge on long-term complications is limited. In particular, a large portion of survivors has respiratory complications, but currently, experienced radiologists and state-of-the-art artificial intelligence systems are not able to detect many abnormalities from follow-up computerized tomography (CT) scans of COVID-19 survivors. Here we propose Deep-LungParenchyma-Enhancing (DLPE), a computer-aided detection (CAD) method for detecting and quantifying pulmonary parenchyma lesions on chest CT. Through proposing a number of deep-learning-based segmentation models and assembling them in an interpretable manner, DLPE removes irrelevant tissues from the perspective of pulmonary parenchyma, and calculates the scan-level optimal window, which considerably enhances parenchyma lesions relative to the lung window. Aided by DLPE, radiologists discovered novel and interpretable lesions from COVID-19 inpatients and survivors, which were previously invisible under the lung window. Based on DLPE, we removed the scan-level bias of CT scans, and then extracted precise radiomics from such novel lesions. We further demonstrated that these radiomics have strong predictive power for key COVID-19 clinical metrics on an inpatient cohort of 1,193 CT scans and for sequelae on a survivor cohort of 219 CT scans. Our work sheds light on the development of interpretable medical artificial intelligence and showcases how artificial intelligence can discover medical findings that are beyond sight.

Computer Aided Automatic Detection and Diagnosis System of Wound and Ulcer Care for Diabetic Patient

The diabetic wound healing process is a complex task under the category of B40 classification and below. The medical expenses are high in private wound specialist organizations compared to government hospitals. This article designed a computer-aided automatic detection and classification method for wound and ulcer care for diabetic patients using image processing techniques by Edge detection, colour scale of tissues, wound area calculation, and percentage calculation with GUI. The system results, Combination of edge detection methodology and 2-D boundary technique and design with the significant three tissues classification which is harmful and required immediate medical responses are Granulation, Fibrin and Necrotic values are used for wound area determination. The result of the system will help the patient immediately, which is classified as high or less severity.

Bollard Segmentation and Position Estimation From Lidar Point Cloud for Autonomous Mooring

This article presents a computer-aided object detection and localization method from lidar 3-D point cloud data. This topic of interest is in the framework of autonomous mooring, where the ship is tied to the rigid structure on-shore (bollard) for autonomous maritime navigation. Using shape and features priors, unlike matching the whole object template to the experimental 3-D point cloud representation of the scene, two customized algorithms: 1) 3-D feature matching (3-DFM) and 2) mixed feature-correspondence matching (MFCM) are presented. The proposed algorithms discriminate and extract the 3-D points corresponding to the noncooperative bollard&#x2019;s surface from the background, thus capable of classification, localization, and representing it using a unique coordinate in the 3-D world. The proposed algorithms are tested and validated by implementing upon an experimental dataset of 105 scenes where the bollard is at different positions and orientations with respect to lidar mounted on the robotic arm. Statistical and probabilistic-based approaches are taken into account to determine the performance of proposed algorithms. Model parameters&#x2019; estimation implies that errors resulting from the 3-DFM algorithm follow homoscedastic bimodal Gaussian distribution with individual Gaussian components having mean 0.03 and 0.09 m, and both have an equal standard deviation of 0.01 m. Furthermore, the posterior component assignment probability is used to identify and cluster the scenes that contribute to relatively larger errors. Finally, an improved algorithm, MFCM, is proposed, whose errors follow unimodal Gaussian distribution with a mean and a standard deviation of 0.03 and 0.01 m, respectively, thus mitigating the shortcomings of the former.

Automatic Detection of Small Intestinal Hookworms in Capsule Endoscopy Images Based on a Convolutional Neural Network.

Ancylostomiasis is a fairly common small bowel parasite disease identified by capsule endoscopy (CE) for which a computer-aided clinical detection method has not been established. We sought to develop an artificial intelligence system with a convolutional neural network (CNN) to automatically detect hookworms in CE images. We trained a deep CNN system based on a YOLO-V4 (You Look Only Once-Version4) detector using 11236 CE images of hookworms. We assessed its performance by calculating the area under the receiver operating characteristic curve and its sensitivity, specificity, and accuracy using an independent test set of 10,529 small-bowel images including 531 images of hookworms. The trained CNN system required 403 seconds to evaluate 10,529 test images. The area under the curve for the detection of hookworms was 0.972 (95% confidence interval (CI), 0.967-0.978). The sensitivity, specificity, and accuracy of the CNN system were 92.2%, 91.1%, and 91.2%, respectively, at a probability score cut-off of 0.485. We developed and validated a CNN-based system for detecting hookworms in CE images. By combining this high-accuracy, high-speed, and oversight-preventing system with other CNN systems, we hope it will become an important supplement for detecting intestinal abnormalities in CE images. This trial is registered with ChiCTR2000034546 (a clinical research of artificial-intelligence-aided diagnosis for hookworms in small intestine by capsule endoscope images).

AUTOMATED POLYP DETECTION IN COLONOSCOPY VIDEOS USING IMAGE ENHANCEMENT AND SALIENCY DETECTION ALGORITHM

Colonoscopy has proven to be an active diagnostic tool that examines the lower half of the digestive system’s anomalies. This paper confers a Computer-Aided Detection (CAD) method for polyps from colonoscopy images that helps to diagnose the early stage of Colorectal Cancer (CRC). The proposed method consists primarily of image enhancement, followed by the creation of a saliency map, feature extraction using the Histogram of Oriented-Gradients (HOG) feature extractor, and classification using the Support Vector Machine (SVM). We present an efficient image enhancement algorithm for highlighting clinically significant features in colonoscopy images. The proposed enhancement approach can improve the overall contrast and brightness by minimizing the effects of inconsistent illumination conditions. Detailed experiments have been conducted using the publicly available colonoscopy databases CVC ClinicDB, CVC ColonDB and the ETIS Larib. The performance measures are found to be in terms of precision (91.69%), recall (81.53%), F1-score (86.31%) and F2-score (89.45%) for the CVC ColonDB database and precision (90.29%), recall (61.73%), F1-score (73.32%) and F2-score (82.64%) for the ETIS Larib database. Comparison with the futuristic method shows that the proposed approach surpasses the existing one in terms of precision, F1-score, and F2-score. The proposed enhancement with saliency-based selection significantly reduced the number of search windows, resulting in an efficient polyp detection algorithm.

Automatic breast lesion detection in ultrafast DCE‐MRI using deep learning

We propose a deep learning-based computer-aided detection (CADe) method to detect breast lesions in ultrafast DCE-MRI sequences. This method uses both the 3D spatial information and temporal information obtained from the early-phase of the dynamic acquisition. The proposed CADe method, based on a modified 3D RetinaNet model, operates on ultrafast T1 weighted sequences, which are preprocessed for motion compensation, temporal normalization, and are cropped before passing into the model. The model is optimized to enable the detection of relatively small breast lesions in a screening setting, focusing on detection of lesions that are harder to differentiate from confounding structures inside the breast. The method was developed based on a dataset consisting of 489 ultrafast MRI studies obtained from 462 patients containing a total of 572 lesions (365 malignant, 207 benign) and achieved a detection rate, sensitivity, and detection rate of benign lesions of 0.90 (0.876-0.934), 0.95 (0.934-0.980), and 0.81 (0.751-0.871) at four false positives per normal breast with 10-fold cross-testing, respectively. The deep learning architecture used for the proposed CADe application can efficiently detect benign and malignant lesions on ultrafast DCE-MRI. Furthermore, utilizing the less visible hard-to-detect lesions in training improves the learning process and, subsequently, detection of malignant breast lesions.

Automatic detection of colorectal neoplasia in wireless colon capsule endoscopic images using a deep convolutional neural network.

Although colorectal neoplasms are the most common abnormalities found in colon capsule endoscopy (CCE), no computer-aided detection method is yet available. We developed an artificial intelligence (AI) system that uses deep learning to automatically detect such lesions in CCE images. We trained a deep convolutional neural network system based on a Single Shot MultiBox Detector using 15 933 CCE images of colorectal neoplasms, such as polyps and cancers. We assessed performance by calculating areas under the receiver operating characteristic curves, along with sensitivities, specificities, and accuracies, using an independent test set of 4784 images, including 1850 images of colorectal neoplasms and 2934 normal colon images. The area under the curve for detection of colorectal neoplasia by the AI model was 0.902. The sensitivity, specificity, and accuracy were 79.0 %, 87.0 %, and 83.9 %, respectively, at a probability cutoff of 0.348. We developed and validated a new AI-based system that automatically detects colorectal neoplasms in CCE images.

Automatic detection of pulmonary nodules on CT images with YOLOv3: development and evaluation using simulated and patient data.

To develop a high-efficiency pulmonary nodule computer-aided detection (CAD) method for localization and diameter estimation. The developed CAD method centralizes a novel convolutional neural network (CNN) algorithm, You Only Look Once (YOLO) v3, as a deep learning approach. This method is featured by two distinct properties: (I) an automatic multi-scale feature extractor for nodule feature screening, and (II) a feature-based bounding box generator for nodule localization and diameter estimation. Two independent studies were performed to train and evaluate this CAD method. One study comprised of a computer simulation that utilized computer-based ground truth. In this study, 300 CT scans were simulated by Cardiac-torso (XCAT) digital phantom. Spherical nodules of various sizes (i.e., 3-10 mm in diameter) were randomly implanted within the lung region of the simulated images-the second study utilized human-based ground truth in patients. The CAD method was developed by CT scans sourced from the LIDC-IDRI database. CT scans with slice thickness above 2.5 mm were excluded, leaving 888 CT images for analysis. A 10-fold cross-validation procedure was implemented in both studies to evaluate network hyper-parameterization and generalization. The overall accuracy of the CAD method was evaluated by the detection sensitivities, in response to average false positives (FPs) per image. In the patient study, the detection accuracy was further compared against 9 recently published CAD studies using free-receiver response operating characteristic (FROC) curve analysis. Localization and diameter estimation accuracies were quantified by the mean and standard error between the predicted value and ground truth. The average results among the 10 cross-validation folds in both studies demonstrated the CAD method achieved high detection accuracy. The sensitivity was 99.3% (FPs =1), and improved to 100% (FPs =4) in the simulation study. The corresponding sensitivities were 90.0% and 95.4% in the patient study, displaying superiority over several conventional and CNN-based lung nodule CAD methods in the FROC curve analysis. Nodule localization and diameter estimation errors were less than 1 mm in both studies. The developed CAD method achieved high computational efficiency: it yields nodule-specific quantitative values (i.e., number, existence confidence, central coordinates, and diameter) within 0.1 s for 2D CT slice inputs. The reported results suggest that the developed lung pulmonary nodule CAD method possesses high accuracies of nodule localization and diameter estimation. The high computational efficiency enables its potential clinical application in the future.

Artificial intelligence using a convolutional neural network for automatic detection of small-bowel angioectasia in capsule endoscopy images.

Although small-bowel angioectasia is reported as the most common cause of bleeding in patients and frequently diagnosed by capsule endoscopy (CE) in patients with obscure gastrointestinal bleeding, a computer-aided detection method has not been established. We developed an artificial intelligence system with deep learning that can automatically detect small-bowel angioectasia in CE images. We trained a deep convolutional neural network (CNN) system based on Single Shot Multibox Detector using 2237 CE images of angioectasia. We assessed its diagnostic accuracy by calculating the area under the receiver operating characteristic curve (ROC-AUC), sensitivity, specificity, positive predictive value, and negative predictive value using an independent test set of 10488 small-bowel images, including 488 images of small-bowel angioectasia. The AUC to detect angioectasia was 0.998. Sensitivity, specificity, positive predictive value, and negative predictive value of CNN were 98.8%, 98.4%, 75.4%, and 99.9%, respectively, at a cut-off value of 0.36 for the probability score. We developed and validated a new system based on CNN to automatically detect angioectasia in CE images. This may be well applicable to daily clinical practice to reduce the burden of physicians as well as to reduce oversight.

Classification of lung nodules using CT images based on texture features and fractal dimension transformation

A new computer-aided detection (CAD) system for lung nodule detection and selection in computed tomography scans is substantiated and implemented. The method consists of the following stages: preprocessing based on threshold and morphological filtration, the formation of suspicious regions of interest using a priori information, the detection of lung nodules by applying the fractal dimension transformation, the computation of informative texture features for identified lung nodules, and their classification by applying the SVM and AdaBoost algorithms. A physical interpretation of the proposed CAD system is given, and its block diagram is constructed. The simulation results based on the proposed CAD method demonstrate advantages of the new approach in terms of standard criteria, such as sensitivity and the false-positive rate.

Classification of Lung Nodules Using CT Images Based on Texture Features and Fractal Dimension Transformation

A new computer-aided detection (CAD) system for lung nodule detection and selection in computed tomography scans is substantiated and implemented. The method consists of the following stages: preprocessing based on threshold and morphological filtration, the formation of suspicious regions of interest using a priori information, the detection of lung nodules by applying the fractal dimension transformation, the computation of informative texture features for identified lung nodules, and their classification by applying the SVM and AdaBoost algorithms. A physical interpretation of the proposed CAD system is given, and its block diagram is constructed. The simulation results based on the proposed CAD method demonstrate advantages of the new approach in terms of standard criteria, such as sensitivity and the false-positive rate.

Multi-channel acoustic analysis of phoneme /s/ mispronunciation for lateral sigmatism detection

The paper presents a method for computer-aided detection of lateral sigmatism. The aim of the study is to design an automated sigmatism diagnosis tool. For that purpose, a reference speech corpus has been collected. It contains 438 recordings of a phoneme /s/ surrounded by certain vowels with normative and simulated pathological pronunciation. The acoustic signal is recorded with an acoustic mask, which is a set of microphones organised in a semi-cylindrical surface around the subject's face. Frames containing /s/ phoneme are subjected to beamforming and feature extraction. Two different feature vectors containing, e.g., Mel-frequency cepstral coefficients and fricative formants, are defined and evaluated in terms of binary classification involving support vector machines. A single-channel analysis is confronted with multi-channel processing. The experimental results show that the multi-channel speech signal processing supported by beamforming is able to increase the pathology detection capabilities in general.

Automatic detection of erosions and ulcerations in wireless capsule endoscopy images based on a deep convolutional neural network

Although erosions and ulcerations are the most common small-bowel abnormalities found on wireless capsule endoscopy (WCE), a computer-aided detection method has not been established. We aimed to develop an artificial intelligence system with deep learning to automatically detect erosions and ulcerations in WCE images. We trained a deep convolutional neural network (CNN) system based on a Single Shot Multibox Detector, using 5360 WCE images of erosions and ulcerations. We assessed its performance by calculating the area under the receiver operating characteristic curve and its sensitivity, specificity, and accuracy using an independent test set of 10,440 small-bowel images including 440 images of erosions and ulcerations. The trained CNN required 233 seconds to evaluate 10,440 test images. The area under the curve for the detection of erosions and ulcerations was 0.958 (95% confidence interval [CI], 0.947-0.968). The sensitivity, specificity, and accuracy of the CNN were 88.2% (95% CI, 84.8%-91.0%), 90.9% (95% CI, 90.3%-91.4%), and 90.8% (95% CI, 90.2%-91.3%), respectively, at a cut-off value of 0.481 for the probability score. We developed and validated a new system based on CNN to automatically detect erosions and ulcerations in WCE images. This may be a crucial step in the development of daily-use diagnostic software for WCE images to help reduce oversights and the burden on physicians.