lesion detection
Recently Published Documents


TOTAL DOCUMENTS

1247
(FIVE YEARS 480)

H-INDEX

53
(FIVE YEARS 12)

2022 ◽  
Vol 3 (1) ◽  
pp. 1-19
Author(s):  
Feng Lu ◽  
Wei Li ◽  
Song Lin ◽  
Chengwangli Peng ◽  
Zhiyong Wang ◽  
...  

Wireless capsule endoscopy is a modern non-invasive Internet of Medical Imaging Things that has been increasingly used in gastrointestinal tract examination. With about one gigabyte image data generated for a patient in each examination, automatic lesion detection is highly desirable to improve the efficiency of the diagnosis process and mitigate human errors. Despite many approaches for lesion detection have been proposed, they mainly focus on large lesions and are not directly applicable to tiny lesions due to the limitations of feature representation. As bleeding lesions are a common symptom in most serious gastrointestinal diseases, detecting tiny bleeding lesions is extremely important for early diagnosis of those diseases, which is highly relevant to the survival, treatment, and expenses of patients. In this article, a method is proposed to extract and fuse multi-scale deep features for detecting and locating both large and tiny lesions. A feature extracting network is first used as our backbone network to extract the basic features from wireless capsule endoscopy images, and then at each layer multiple regions could be identified as potential lesions. As a result, the features maps of those potential lesions are obtained at each level and fused in a top-down manner to the fully connected layer for producing final detection results. Our proposed method has been evaluated on a clinical dataset that contains 20,000 wireless capsule endoscopy images with clinical annotation. Experimental results demonstrate that our method can achieve 98.9% prediction accuracy and 93.5% score, which has a significant performance improvement of up to 31.69% and 22.12% in terms of recall rate and score, respectively, when compared to the state-of-the-art approaches for both large and tiny bleeding lesions. Moreover, our model also has the highest AP and the best medical diagnosis performance compared to state-of-the-art multi-scale models.


2022 ◽  
Vol 27 (1) ◽  
pp. 103-113
Author(s):  
Jiacheng Jiao ◽  
Haiwei Pan ◽  
Chunling Chen ◽  
Tao Jin ◽  
Yang Dong ◽  
...  

Cancers ◽  
2022 ◽  
Vol 14 (2) ◽  
pp. 376
Author(s):  
Natália Alves ◽  
Megan Schuurmans ◽  
Geke Litjens ◽  
Joeran S. Bosma ◽  
John Hermans ◽  
...  

Early detection improves prognosis in pancreatic ductal adenocarcinoma (PDAC), but is challenging as lesions are often small and poorly defined on contrast-enhanced computed tomography scans (CE-CT). Deep learning can facilitate PDAC diagnosis; however, current models still fail to identify small (<2 cm) lesions. In this study, state-of-the-art deep learning models were used to develop an automatic framework for PDAC detection, focusing on small lesions. Additionally, the impact of integrating the surrounding anatomy was investigated. CE-CT scans from a cohort of 119 pathology-proven PDAC patients and a cohort of 123 patients without PDAC were used to train a nnUnet for automatic lesion detection and segmentation (nnUnet_T). Two additional nnUnets were trained to investigate the impact of anatomy integration: (1) segmenting the pancreas and tumor (nnUnet_TP), and (2) segmenting the pancreas, tumor, and multiple surrounding anatomical structures (nnUnet_MS). An external, publicly available test set was used to compare the performance of the three networks. The nnUnet_MS achieved the best performance, with an area under the receiver operating characteristic curve of 0.91 for the whole test set and 0.88 for tumors <2 cm, showing that state-of-the-art deep learning can detect small PDAC and benefits from anatomy information.


2022 ◽  
Vol 22 (1) ◽  
Author(s):  
Anying Bai ◽  
Jiaxu Wang ◽  
Qing Li ◽  
Samuel Seery ◽  
Peng Xue ◽  
...  

Abstract Background Inappropriate management of high-grade squamous intraepithelial lesions (HSIL) may be the result of an inaccurate colposcopic diagnosis. The aim of this study was to assess colposcopic performance in identifying HSIL+ cases and to analyze the associated clinical factors. Methods Records from 1130 patients admitted to Shenzhen Maternal and Child Healthcare Hospital from 12th January, 2018 up until 30th December, 2018 were retrospectively collected, and included demographics, cytological results, HPV status, transformation zone type, number of cervical biopsy sites, colposcopists’ competencies, colposcopic impressions, as well as histopathological results. Colposcopy was carried out using 2011 colposcopic terminology from the International Federation of Cervical Pathology and Colposcopy. Logistic regression modelling was implemented for uni- and multivariate analyses. A forward stepwise approach was adopted in order to identify variables associated with colposcopic accuracy. Histopathologic results were taken as the comparative gold standard. Results Data from 1130 patient records were collated and analyzed. Colposcopy was 69.7% accurate in identifying HSIL+ cases. Positive predictive value, negative predictive value, sensitivity and specificity of detecting HSIL or more (HSIL+) were 35.53%, 64.47%, 42.35% and 77.60%, respectively. Multivariate analysis highlighted the number of biopsies, cytology, and transformation zone type as independent factors. Age and HPV subtype did not appear to statistically correlate with high-grade lesion/carcinoma. Conclusion Evidence presented here suggests that colposcopy is only 69.7% accurate at diagnosing HSIL. Even though not all HSIL will progress into cancer it is considered pre-cancerous and therefore early identification will save lives. The number of biopsies, cytology and transformation zone type appear to be predictors of misdiagnosis and therefore should be considered during clinical consultations and by way of further research.


Diagnostics ◽  
2022 ◽  
Vol 12 (1) ◽  
pp. 154
Author(s):  
Joo Hye Song ◽  
Ji Eun Kim ◽  
Hwe Hoon Chung ◽  
Sung Noh Hong ◽  
Heejung Kim ◽  
...  

Video capsule endoscopy (VCE) has become the noninvasive diagnostic standard in the investigation of overt obscure gastrointestinal bleeding (OGIB), with a high positive and negative predictive value. However, the diagnostic yield of the VCE is thought to depend on when it was performed. We evaluate the optimal timing performing VCE relative to overt OGIB to improve the diagnostic yield. A total 271 patients had admitted and underwent VCE for overt OGIB between 2007 and 2016 in Samsung Medical Center, Seoul, Korea. To evaluate the diagnostic yield of VCE for overt OGIB with respect to timing of the intervention, diagnostic yield was analyzed according to the times after latest bleeding. The finding of VCE was classified into P0 or P1 (no potential for bleeding or uncertain hemorrhagic potential) and P2 (high potential for bleeding, such as active bleeding, typical angiodysplasia, large ulcerations or tumors). The P2 lesion was found in 106 patients and diagnostic yield of was 39.1% for overt OGIB. Diagnostic yield of VCE to detect P2 lesion was higher when it is performed closer to the time of latest bleeding (timing of VCE between the VCE and latest bleeding: <24 h, 43/63 (68.3%); 1 days, 16/43 (34.9%); 2 days, 18/52 (34.6%); 3 days, 13/43 (30.2%); 4 days, 7/28 (25.0%); 5–7 days, 6/24 (25.0%), and ≥8 days, 4/18 (22.2%); ptrend <0.001). The interval between the VCE and latest bleeding were categorized into <24 h (n = 63), 1–2 days (n = 95), 3–7 days (n = 95) and ≥8 days (n = 18). Multivariable analyses showed the odds ratio for P2 lesion detection was 4.99 (95% confidence interval, 1.47–16.89) in <24 h group, compared with ≥8 days group (p < 0.010). The overall re-bleeding rate for those with P2 lesion was higher than for those with P0 or P1 lesion at the end of mean follow up of 2.5 years. The proportion of patients who underwent therapeutic intervention including surgery, endoscopic intervention and embolization was higher when VCE is performed closer to the time of latest bleeding (p = 0.010). Early deployment of VCE within 24 h of the latest GI bleeding results in a higher diagnostic yield for patients with overt OGIB and consequently resulted in a higher therapeutic intervention rate.


Diagnostics ◽  
2022 ◽  
Vol 12 (1) ◽  
pp. 148
Author(s):  
Kyungsoo Bae ◽  
Kyung Nyeo Jeon ◽  
Moon Jung Hwang ◽  
Yunsub Jung ◽  
Joonsung Lee

(1) Background: Highly flexible adaptive image receive (AIR) coil has become available for clinical use. The present study aimed to evaluate the performance of AIR anterior array coil in lung MR imaging using a zero echo time (ZTE) sequence compared with conventional anterior array (CAA) coil. (2) Methods: Sixty-six patients who underwent lung MR imaging using both AIR coil (ZTE-AIR) and CAA coil (ZTE-CAA) were enrolled. Image quality of ZTE-AIR and ZTE-CAA was quantified by calculating blur metric value, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) of lung parenchyma. Image quality was qualitatively assessed by two independent radiologists. Lesion detection capabilities for lung nodules and emphysema and/or lung cysts were evaluated. Patients’ comfort levels during examinations were assessed. (3) Results: SNR and CNR of lung parenchyma were higher (both p < 0.001) in ZTE-AIR than in ZTE-CAA. Image sharpness was superior in ZTE-AIR (p < 0.001). Subjective image quality assessed by two independent readers was superior (all p < 0.05) in ZTE-AIR. AIR coil was preferred by 64 of 66 patients. ZTE-AIR showed higher (all p < 0.05) sensitivity for sub-centimeter nodules than ZTE-CAA by both readers. ZTE-AIR showed higher (all p < 0.05) sensitivity and accuracy for detecting emphysema and/or cysts than ZTE-CAA by both readers. (4) Conclusions: The use of highly flexible AIR coil in ZTE lung MR imaging can improve image quality and patient comfort. Application of AIR coil in parenchymal imaging has potential for improving delineation of low-density parenchymal lesions and tiny nodules.


Cancers ◽  
2022 ◽  
Vol 14 (2) ◽  
pp. 277
Author(s):  
Zuzanna Anna Magnuska ◽  
Benjamin Theek ◽  
Milita Darguzyte ◽  
Moritz Palmowski ◽  
Elmar Stickeler ◽  
...  

Automation of medical data analysis is an important topic in modern cancer diagnostics, aiming at robust and reproducible workflows. Therefore, we used a dataset of breast US images (252 malignant and 253 benign cases) to realize and compare different strategies for CAD support in lesion detection and classification. Eight different datasets (including pre-processed and spatially augmented images) were prepared, and machine learning algorithms (i.e., Viola–Jones; YOLOv3) were trained for lesion detection. The radiomics signature (RS) was derived from detection boxes and compared with RS derived from manually obtained segments. Finally, the classification model was established and evaluated concerning accuracy, sensitivity, specificity, and area under the Receiver Operating Characteristic curve. After training on a dataset including logarithmic derivatives of US images, we found that YOLOv3 obtains better results in breast lesion detection (IoU: 0.544 ± 0.081; LE: 0.171 ± 0.009) than the Viola–Jones framework (IoU: 0.399 ± 0.054; LE: 0.096 ± 0.016). Interestingly, our findings show that the classification model trained with RS derived from detection boxes and the model based on the RS derived from a gold standard manual segmentation are comparable (p-value = 0.071). Thus, deriving radiomics signatures from the detection box is a promising technique for building a breast lesion classification model, and may reduce the need for the lesion segmentation step in the future design of CAD systems.


2022 ◽  
pp. 028418512110701
Author(s):  
Jonas Oppenheimer ◽  
Keno Kyrill Bressem ◽  
Fabian Henry Jürgen Elsholtz ◽  
Bernd Hamm ◽  
Stefan Markus Niehues

Background Computed tomography is a standard imaging procedure for the detection of liver lesions, such as metastases, which can often be small and poorly contrasted, and therefore hard to detect. Advances in image reconstruction have shown promise in reducing image noise and improving low-contrast detectability. Purpose To examine a novel, specialized, model-based iterative reconstruction (MBIR) technique for improved low-contrast liver lesion detection. Material and Methods Patient images with reported poorly contrasted focal liver lesions were retrospectively reconstructed with the low-contrast attenuating algorithm (FIRST-LCD) from primary raw data. Liver-to-lesion contrast, signal-to-noise, and contrast-to-noise ratios for background and liver noise for each lesion were compared for all three FIRST-LCD presets with the established hybrid iterative reconstruction method (AIDR-3D). An additional visual conspicuity score was given by two experienced radiologists for each lesion. Results A total of 82 lesions in 57 examinations were included in the analysis. All three FIRST-LCD algorithms provided statistically significant increases in liver-to-lesion contrast, with FIRSTMILD showing the largest increase (40.47 HU in AIDR-3D; 45.84 HU in FIRSTMILD; P < 0.001). Substantial improvement was shown in contrast-to-noise metrics. Visual analysis of the lesions shows decreased lesion visibility with all FIRST methods in comparison to AIDR-3D, with FIRSTSTR showing the closest results ( P < 0.001). Conclusion Objective image metrics show promise for MBIR methods in improving the detectability of low-contrast liver lesions; however, subjective image quality may be perceived as inferior. Further improvements are necessary to enhance image quality and lesion detection.


2022 ◽  
Author(s):  
R. H. G. J. van Lanen ◽  
C. J. Wiggins ◽  
A. J. Colon ◽  
W. H. Backes ◽  
J. F. A. Jansen ◽  
...  

Abstract Purpose Resective epilepsy surgery is a well-established, evidence-based treatment option in patients with drug-resistant focal epilepsy. A major predictive factor of good surgical outcome is visualization and delineation of a potential epileptogenic lesion by MRI. However, frequently, these lesions are subtle and may escape detection by conventional MRI (≤ 3 T). Methods We present the EpiUltraStudy protocol to address the hypothesis that application of ultra-high field (UHF) MRI increases the rate of detection of structural lesions and functional brain aberrances in patients with drug-resistant focal epilepsy who are candidates for resective epilepsy surgery. Additionally, therapeutic gain will be addressed, testing whether increased lesion detection and tailored resections result in higher rates of seizure freedom 1 year after epilepsy surgery. Sixty patients enroll the study according to the following inclusion criteria: aged ≥ 12 years, diagnosed with drug-resistant focal epilepsy with a suspected epileptogenic focus, negative conventional 3 T MRI during pre-surgical work-up. Results All patients will be evaluated by 7 T MRI; ten patients will undergo an additional 9.4 T MRI exam. Images will be evaluated independently by two neuroradiologists and a neurologist or neurosurgeon. Clinical and UHF MRI will be discussed in the multidisciplinary epilepsy surgery conference. Demographic and epilepsy characteristics, along with postoperative seizure outcome and histopathological evaluation, will be recorded. Conclusion This protocol was reviewed and approved by the local Institutional Review Board and complies with the Declaration of Helsinki and principles of Good Clinical Practice. Results will be submitted to international peer-reviewed journals and presented at international conferences. Trial registration number www.trialregister.nl: NTR7536.


Sign in / Sign up

Export Citation Format

Share Document