Computer-Aided Detection of Malignant Lung Nodules on Chest Radiographs: Effect on Observers’ Performance

이경희,구진모,박창민,이현주,진광남 대한영상의학회 2012 Korean Journal of Radiology Vol.13 No.5

Objective: To evaluate the effect of computer-aided detection (CAD) system on observer performance in the detection of malignant lung nodules on chest radiograph. Materials and Methods: Two hundred chest radiographs (100 normal and 100 abnormal with malignant solitary lung nodules) were evaluated. With CT and histological confirmation serving as a reference, the mean nodule size was 15.4 mm (range, 7-20 mm). Five chest radiologists and five radiology residents independently interpreted both the original radiographs and CAD output images using the sequential testing method. The performances of the observers for the detection of malignant nodules with and without CAD were compared using the jackknife free-response receiver operating characteristic analysis. Results: Fifty-nine nodules were detected by the CAD system with a false positive rate of 1.9 nodules per case. The detection of malignant lung nodules significantly increased from 0.90 to 0.92 for a group of observers, excluding one first-year resident (p = 0.04). When lowering the confidence score was not allowed, the average figure of merit also increased from 0.90 to 0.91 (p = 0.04) for all observers after a CAD review. On average, the sensitivities with and without CAD were 87% and 84%, respectively; the false positive rates per case with and without CAD were 0.19 and 0.17, respectively. The number of additional malignancies detected following true positive CAD marks ranged from zero to seven for the various observers. Conclusion: The CAD system may help improve observer performance in detecting malignant lung nodules on chest radiographs and contribute to a decrease in missed lung cancer. Objective: To evaluate the effect of computer-aided detection (CAD) system on observer performance in the detection of malignant lung nodules on chest radiograph. Materials and Methods: Two hundred chest radiographs (100 normal and 100 abnormal with malignant solitary lung nodules) were evaluated. With CT and histological confirmation serving as a reference, the mean nodule size was 15.4 mm (range, 7-20 mm). Five chest radiologists and five radiology residents independently interpreted both the original radiographs and CAD output images using the sequential testing method. The performances of the observers for the detection of malignant nodules with and without CAD were compared using the jackknife free-response receiver operating characteristic analysis. Results: Fifty-nine nodules were detected by the CAD system with a false positive rate of 1.9 nodules per case. The detection of malignant lung nodules significantly increased from 0.90 to 0.92 for a group of observers, excluding one first-year resident (p = 0.04). When lowering the confidence score was not allowed, the average figure of merit also increased from 0.90 to 0.91 (p = 0.04) for all observers after a CAD review. On average, the sensitivities with and without CAD were 87% and 84%, respectively; the false positive rates per case with and without CAD were 0.19 and 0.17, respectively. The number of additional malignancies detected following true positive CAD marks ranged from zero to seven for the various observers. Conclusion: The CAD system may help improve observer performance in detecting malignant lung nodules on chest radiographs and contribute to a decrease in missed lung cancer.

Computer-Aided Detection of Malignant Lung Nodules on Chest Radiographs: Effect on Observers' Performance

Lee, Kyung Hee,Goo, Jin Mo,Park, Chang Min,Lee, Hyun Ju,Jin, Kwang Nam The Korean Society of Radiology 2012 KOREAN JOURNAL OF RADIOLOGY Vol.13 No.5

<P><B>Objective</B></P><P>To evaluate the effect of computer-aided detection (CAD) system on observer performance in the detection of malignant lung nodules on chest radiograph.</P><P><B>Materials and Methods</B></P><P>Two hundred chest radiographs (100 normal and 100 abnormal with malignant solitary lung nodules) were evaluated. With CT and histological confirmation serving as a reference, the mean nodule size was 15.4 mm (range, 7-20 mm). Five chest radiologists and five radiology residents independently interpreted both the original radiographs and CAD output images using the sequential testing method. The performances of the observers for the detection of malignant nodules with and without CAD were compared using the jackknife free-response receiver operating characteristic analysis.</P><P><B>Results</B></P><P>Fifty-nine nodules were detected by the CAD system with a false positive rate of 1.9 nodules per case. The detection of malignant lung nodules significantly increased from 0.90 to 0.92 for a group of observers, excluding one first-year resident (<I>p</I> = 0.04). When lowering the confidence score was not allowed, the average figure of merit also increased from 0.90 to 0.91 (<I>p</I> = 0.04) for all observers after a CAD review. On average, the sensitivities with and without CAD were 87% and 84%, respectively; the false positive rates per case with and without CAD were 0.19 and 0.17, respectively. The number of additional malignancies detected following true positive CAD marks ranged from zero to seven for the various observers.</P><P><B>Conclusion</B></P><P>The CAD system may help improve observer performance in detecting malignant lung nodules on chest radiographs and contribute to a decrease in missed lung cancer.</P>

Computer-Aided Detection with Automated Breast Ultrasonography for Suspicious Lesions Detected on Breast MRI

Kim, Sanghee,Kang, Bong Joo,Kim, Sung Hun,Lee, Jeongmin,Park, Ga Eun Korean Society of Magnetic Resonance in Medicine 2019 Investigative Magnetic Resonance Imaging Vol.23 No.1

Purpose: The aim of this study was to evaluate the diagnostic performance of a computer-aided detection (CAD) system used with automated breast ultrasonography (ABUS) for suspicious lesions detected on breast MRI, and CAD-false lesions. Materials and Methods: We included a total of 40 patients diagnosed with breast cancer who underwent ABUS (ACUSON S2000) to evaluate multiple suspicious lesions found on MRI. We used CAD ($QVCAD^{TM}$) in all the ABUS examinations. We evaluated the diagnostic accuracy of CAD and analyzed the characteristics of CAD-detected lesions and the factors underlying false-positive and false-negative cases. We also analyzed false-positive lesions with CAD on ABUS. Results: Of a total of 122 suspicious lesions detected on MRI in 40 patients, we excluded 51 daughter nodules near the main breast cancer within the same quadrant and included 71 lesions. We also analyzed 23 false-positive lesions using CAD with ABUS. The sensitivity, specificity, positive predictive value, and negative predictive value of CAD (for 94 lesions) with ABUS were 75.5%, 44.4%, 59.7%, and 62.5%, respectively. CAD facilitated the detection of 81.4% (35/43) of the invasive ductal cancer and 84.9% (28/33) of the invasive ductal cancer that showed a mass (excluding non-mass). CAD also revealed 90.3% (28/31) of the invasive ductal cancers measuring larger than 1 cm (excluding non-mass and those less than 1 cm). The mean sizes of the true-positive versus false-negative mass lesions were $2.08{\pm}0.85cm$ versus $1.6{\pm}1.28cm$ (P < 0.05). False-positive lesions included sclerosing adenosis and usual ductal hyperplasia. In a total of 23 false cases of CAD, the most common (18/23) cause was marginal or subareolar shadowing, followed by three simple cysts, a hematoma, and a skin wart. Conclusion: CAD with ABUS showed promising sensitivity for the detection of invasive ductal cancer showing masses larger than 1 cm on MRI.

Single-view 2D CNNs with fully automatic non-nodule categorization for false positive reduction in pulmonary nodule detection

Eun, Hyunjun,Kim, Daeyeong,Jung, Chanho,Kim, Changick Elsevier 2018 Computer methods and programs in biomedicine Vol.165 No.-

<P><B>Abstract</B></P> <P>Background and Objective: In pulmonary nodule detection, the first stage, candidate detection, aims to detect suspicious pulmonary nodules. However, detected candidates include many false positives and thus in the following stage, false positive reduction, such false positives are reliably reduced. Note that this task is challenging due to 1) the imbalance between the numbers of nodules and non-nodules and 2) the intra-class diversity of non-nodules. Although techniques using 3D convolutional neural networks (CNNs) have shown promising performance, they suffer from high computational complexity which hinders constructing deep networks. To efficiently address these problems, we propose a novel framework using the ensemble of 2D CNNs using single views, which outperforms existing 3D CNN-based methods.</P> <P>Methods: Our ensemble of 2D CNNs utilizes single-view 2D patches to improve both computational and memory efficiency compared to previous techniques exploiting 3D CNNs. We first categorize non-nodules on the basis of features encoded by an autoencoder. Then, all 2D CNNs are trained by using the same nodule samples, but with different types of non-nodules. By extending the learning capability, this training scheme resolves difficulties of extracting representative features from non-nodules with large appearance variations. Note that, instead of manual categorization requiring the heavy workload of radiologists, we propose to automatically categorize non-nodules based on the autoencoder and k-means clustering.</P> <P>Results: We performed extensive experiments to validate the effectiveness of our framework based on the database of the lung nodule analysis 2016 challenge. The superiority of our framework is demonstrated through comparing the performance of five frameworks trained with differently constructed training sets. Our proposed framework achieved state-of-the-art performance (0.922 of the competition performance metric score) with low computational demands (789K of parameters and 1024M of floating point operations per second).</P> <P>Conclusion: We presented a novel false positive reduction framework, the ensemble of single-view 2D CNNs with fully automatic non-nodule categorization, for pulmonary nodule detection. Unlike previous 3D CNN-based frameworks, we utilized 2D CNNs using 2D single views to improve computational efficiency. Also, our training scheme using categorized non-nodules, extends the learning capability of representative features of different non-nodules. Our framework achieved state-of-the-art performance with low computational complexity.</P> <P><B>Highlights</B></P> <P> <UL> <LI> We propose a novel framework, an ensemble of 2D CNNs using single views, for efficient and accurate false positive reduction in pulmonary nodule detection. </LI> <LI> We introduce a fully automatic non-nodule categorization by utilizing an autoencoder and k-means clustering to extend the learning capability of our network. </LI> <LI> The proposed framework utilizes 2D patches to improve memory usage and computational efficiency without a decrease in performance. </LI> </UL> </P>

Computer-Aided Detection with Automated Breast Ultrasonography for Suspicious Lesions Detected on Breast MRI

김상희,강봉주,김성헌,이정민,박가은 대한자기공명의과학회 2019 Investigative Magnetic Resonance Imaging Vol.23 No.1

Purpose: The aim of this study was to evaluate the diagnostic performance of a computer-aided detection (CAD) system used with automated breast ultrasonography (ABUS) for suspicious lesions detected on breast MRI, and CAD-false lesions. Materials and Methods: We included a total of 40 patients diagnosed with breast cancer who underwent ABUS (ACUSON S2000) to evaluate multiple suspicious lesions found on MRI. We used CAD (QVCADTM) in all the ABUS examinations. We evaluated the diagnostic accuracy of CAD and analyzed the characteristics of CAD-detected lesions and the factors underlying false-positive and false-negative cases. We also analyzed false-positive lesions with CAD on ABUS. Results: Of a total of 122 suspicious lesions detected on MRI in 40 patients, we excluded 51 daughter nodules near the main breast cancer within the same quadrant and included 71 lesions. We also analyzed 23 false-positive lesions using CAD with ABUS. The sensitivity, specificity, positive predictive value, and negative predictive value of CAD (for 94 lesions) with ABUS were 75.5%, 44.4%, 59.7%, and 62.5%, respectively. CAD facilitated the detection of 81.4% (35/43) of the invasive ductal cancer and 84.9% (28/33) of the invasive ductal cancer that showed a mass (excluding non-mass). CAD also revealed 90.3% (28/31) of the invasive ductal cancers measuring larger than 1 cm (excluding non-mass and those less than 1 cm). The mean sizes of the true-positive versus false-negative mass lesions were 2.08 ± 0.85 cm versus 1.6 ± 1.28 cm (P < 0.05). False-positive lesions included sclerosing adenosis and usual ductal hyperplasia. In a total of 23 false cases of CAD, the most common (18/23) cause was marginal or subareolar shadowing, followed by three simple cysts, a hematoma, and a skin wart. Conclusion: CAD with ABUS showed promising sensitivity for the detection of invasive ductal cancer showing masses larger than 1 cm on MRI.

Simultaneous detection and classification of breast masses in digital mammograms via a deep learning YOLO-based CAD system

Al-masni, Mohammed A.,Al-antari, Mugahed A.,Park, Jeong-Min,Gi, Geon,Kim, Tae-Yeon,Rivera, Patricio,Valarezo, Edwin,Choi, Mun-Taek,Han, Seung-Moo,Kim, Tae-Seong Elsevier 2018 COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE Vol.157 No.-

<P><B>Abstract</B></P> <P><B>Background and objective</B></P> <P>Automatic detection and classification of the masses in mammograms are still a big challenge and play a crucial role to assist radiologists for accurate diagnosis. In this paper, we propose a novel Computer-Aided Diagnosis (CAD) system based on one of the regional deep learning techniques, a ROI-based Convolutional Neural Network (CNN) which is called You Only Look Once (YOLO). Although most previous studies only deal with classification of masses, our proposed YOLO-based CAD system can handle detection and classification simultaneously in one framework.</P> <P><B>Methods</B></P> <P>The proposed CAD system contains four main stages: preprocessing of mammograms, feature extraction utilizing deep convolutional networks, mass detection with confidence, and finally mass classification using Fully Connected Neural Networks (FC-NNs). In this study, we utilized original 600 mammograms from Digital Database for Screening Mammography (DDSM) and their augmented mammograms of 2,400 with the information of the masses and their types in training and testing our CAD. The trained YOLO-based CAD system detects the masses and then classifies their types into benign or malignant.</P> <P><B>Results</B></P> <P>Our results with five-fold cross validation tests show that the proposed CAD system detects the mass location with an overall accuracy of 99.7%. The system also distinguishes between benign and malignant lesions with an overall accuracy of 97%.</P> <P><B>Conclusions</B></P> <P>Our proposed system even works on some challenging breast cancer cases where the masses exist over the pectoral muscles or dense regions.</P> <P><B>Highlights</B></P> <P> <UL> <LI> A novel computer-aided diagnosis system based on deep learning techniques is proposed. </LI> <LI> The proposed YOLO-based CAD system simultaneously handles both detection and classification of breast cancer masses. </LI> <LI> YOLO-based CAD has a capability to handle most challenging cases of breast abnormalities. </LI> </UL> </P>

Does computer-aided diagnostic endoscopy improve the detection of commonly missed polyps? A meta-analysis

Arun Sivananthan,Scarlet Nazarian,Lakshmana Ayaru,Kinesh Patel,Hutan Ashrafian,Ara Darzi,Nisha Patel 대한소화기내시경학회 2022 Clinical Endoscopy Vol.55 No.3

Background/Aims: Colonoscopy is the gold standard diagnostic method for colorectal neoplasia, allowing detection and resection ofadenomatous polyps; however, significant proportions of adenomas are missed. Computer-aided detection (CADe) systems in endoscopyare currently available to help identify lesions. Diminutive (≤5 mm) and nonpedunculated polyps are most commonly missed. This meta-analysis aimed to assess whether CADe systems can improve the real-time detection of these commonly missed lesions. Methods: A comprehensive literature search was performed. Randomized controlled trials evaluating CADe systems categorized bymorphology and lesion size were included. The mean number of polyps and adenomas per patient was derived. Independent proportionsand their differences were calculated using DerSimonian and Laird random-effects modeling. Results: Seven studies, including 2,595 CADe-assisted colonoscopies and 2,622 conventional colonoscopies, were analyzed. CADe-assistedcolonoscopy demonstrated an 80% increase in the mean number of diminutive adenomas detected per patient compared withconventional colonoscopy (0.31 vs. 0.17; effect size, 0.13; 95% confidence interval [CI], 0.09−0.18); it also demonstrated a 91.7% increasein the mean number of nonpedunculated adenomas detected per patient (0.32 vs. 0.19; effect size, 0.05; 95% CI, 0.02−0.07). Conclusions: CADe-assisted endoscopy significantly improved the detection of most commonly missed adenomas. Although thismethod is a potentially exciting technology, limitations still apply to current data, prompting the need for further real-time studies.

유방영상의학과 인공지능: 현재와 미래

김희정,김학희 대한가정의학회 2023 Korean Journal of Family Practice Vol.13 No.4

유방암은 전 세계 여성에서 가장 흔한 암으로, 높은 발병률과 사망률을 고려할 때 조기 발견이 매우 중요하다. 현재까지 유방촬영술이 유방암의 주요 검진 기법으로 사용되고 있지만, 유방촬영술은 위양성 결과를 보일 수 있고, 치밀 유방에서는 병변을 발견하는 데 어려움이 있는 등 여러 제한점이 있다. 이를 극복하기 위해 유방암을 의심할만한 패턴을 식별하는 인공지능 기반 컴퓨터 발견 보조(artificial intelligence-based computer-aided detection, AI-CAD) 시스템이 개발되었다. 여러 연구에서 AI-CAD는 유방암을 높은 민감도와 특이도로 검출하였고, 영상의학과 의사보다 높은 성능을 보이기도 했다. 더불어 AI-CAD는 영상의학과 의사의 판독 성적을 유의하게 향상시키는 것으로 나타났고, 특히 판독의 난이도가 높은 케이스에서 진단 정확도가 향상되는 것으로 나타났다. AI는 유방암 검출뿐만 아니라 업무 흐름 최적화, 영상 판독의 우선 순위 설정, 유방 밀도 평가, 및 개별 유방암 위험 예측에도 이용될 수 있다. 하지만 여전히 AI-CAD 의 일반화 가능성, 일관성, 및 효율성을 보장할 수 있는 대규모 임상 검증이 필요하며, 임상 적용을 위한 윤리적, 법적 고찰 또한 필요하다. 나아가 AI 기술의 잠재력을 극대화하기 위해서는 영상의학과 의사와 AI의 역할을 명확히 정의하고 협력하도록 해야 한다. 결론적으로, AI는 유방암 검출을 촉진하고 영상의학과 의사의 판독 성적을 향상시키는 데 높은 가능성을 보여 주었으며, 향후 널리 도입되기 위해서는 추가 연구와 임상 검증, 윤리적 고려, 그리고 영상의학과 의사와 AI 간의 협력이 필요하다. Breast cancer stands as a pervasive and life-threatening disease affecting women worldwide. Early detection of the disease is critical, and mammography has conventionally been a primary screening modality. However, mammography has some limitations, such as false-positives and difficulties in detecting lesions in dense breasts. To address these challenges, artificial intelligence-based computer-aided detection (AI-CAD) systems have been developed to assist radiologists in interpreting mammograms. These systems identify suspicious patterns indicative of breast cancer. Studies have shown that AI-CAD exhibits high sensitivity and specificity in breast cancer detection, at times even surpassing the performance of radiologists. Radiologists utilizing AI-CAD achieved enhanced diagnostic accuracy, especially in complex cases. Notably, AI’s role extends beyond cancer detection. It can streamline workflow, help radiologists prioritize images, objectively evaluate breast density, and predict individual breast cancer risk. While AI shows promise in breast imaging, challenges still persist. Large-scale clinical validation is needed to establish generalizability, consistency, and efficiency. Ethical and legal considerations are essential for integrating AI into clinical practice. Collaborations between radiologists and AI is important for maximizing the technology’s potential in breast imaging. In conclusion, AI has shown significant promise in enhancing breast cancer detection and improving radiologists’ performance in breast imaging. Further research, clinical validation, ethical considerations, and collaborations between radiologists and AI are crucial for its widespread adoption in breast imaging.

A Tuberculosis Detection Method Using Attention and Sparse R-CNN

Xuebin Xu,Jiada Zhang,Xiaorui Cheng,Longbin Lu,Yuqing Zhao,Zongyu Xu,Zhuangzhuang Gu 한국인터넷정보학회 2022 KSII Transactions on Internet and Information Syst Vol.16 No.7

To achieve accurate detection of tuberculosis (TB) areas in chest radiographs, we design a chest X-ray TB area detection algorithm. The algorithm consists of two stages: the chest X-ray TB classification network (CXTCNet) and the chest X-ray TB area detection network (CXTDNet). CXTCNet is used to judge the presence or absence of TB areas in chest X-ray images, thereby excluding the influence of other lung diseases on the detection of TB areas. It can reduce false positives in the detection network and improve the accuracy of detection results. In CXTCNet, we propose a channel attention mechanism (CAM) module and combine it with DenseNet. This module enables the network to learn more spatial and channel features information about chest X-ray images, thereby improving network performance. CXTDNet is a design based on a sparse object detection algorithm (Sparse R-CNN). A group of fixed learnable proposal boxes and learnable proposal features are using for classification and location. The predictions of the algorithm are output directly without non-maximal suppression post-processing. Furthermore, we use CLAHE to reduce image noise and improve image quality for data preprocessing. Experiments on dataset TBX11K show that the accuracy of the proposed CXTCNet is up to 99.10%, which is better than most current TB classification algorithms. Finally, our proposed chest X-ray TB detection algorithm could achieve AP of 45.35% and AP50 of 74.20%. We also establish a chest X-ray TB dataset with 304 sheets. And experiments on this dataset showed that the accuracy of the diagnosis was comparable to that of radiologists. We hope that our proposed algorithm and established dataset will advance the field of TB detection.

Usefulness of the CAD System for Detecting Pulmonary Nodule in Real Clinical Practice

송경두,정명진,김희철,정선영,이경수 대한영상의학회 2011 Korean Journal of Radiology Vol.12 No.2

Objective: We wanted to evaluate the usefulness of the computer-aided detection (CAD) system for detecting pulmonary nodules in real clinical practice by using the CT images. Materials and Methods: Our Institutional Review Board approved our retrospective study with a waiver of informed consent. This study included 166 CT examinations that were performed for the evaluation of pulmonary metastasis in 166 patients with colorectal cancer. All the CT examinations were interpreted by radiologists and they were also evaluated by the CAD system. All the nodules detected by the CAD system were evaluated with regard to whether or not they were true nodules, and they were classifi ed into micronodules (MN, diameter < 4 mm) and signifi cant nodules (SN, 4 ≤ diameter ≤ 10 mm). The radiologic reports and CAD results were compared. Results: The CAD system helped detect 426 nodules; 115 (27%) of the 426 nodules were classifi ed as true nodules and 35 (30%) of the 115 nodules were SNs, and 83 (72%) of the 115 were not mentioned in the radiologists’ reports and three (4%) of the 83 nodules were non-calcifi ed SNs. One of three non-calcifi ed SNs was confi rmed as a metastatic nodule. According to the radiologists’ reports, 60 true nodules were detected, and 28 of the 60 were not detected by the CAD system. Conclusion: Although the CAD system missed many SNs that are detected by radiologists, it helps detect additional nodules that are missed by the radiologists in real clinical practice. Therefore, the CAD system can be useful to support a radiologist’s detection performance.