Detection Of Lesions Research Articles

The current infection with Helicobacter pylori and association with upper gastrointestinal lesions and risk of upper gastrointestinal cancer: Insights from multicenter population-based cohort study.

The relationship between Helicobacter pylori (H. pylori) infection and upper gastrointestinal (UGI) cancers is complex. This multicenter, population-based cohort study conducted in seven areas in China aimed to assess the correlation between current H. pylori infection and the severity of UGI lesions, as well as its association with the risk of gastric cancer (GC) and esophageal cancer (EC). From 2015 to 2017, 27,085 participants (aged 40-69) completed a standardized questionnaire, and underwent a 13C-urea breath test. Then a subset underwent UGI endoscopy to assess the UGI lesion detection rates. All individuals were followed up until December 2021 to calculate the hazard ratios (HRs) for UGI cancers. H. pylori infection prevalence was 45.9%, and among endoscopy participants, 22.2% had gastric lesions, 19.2% had esophageal lesions. Higher detection rates of gastric lesions were noted in the H. pylori-positive population across all lesion severity levels. Over a median follow-up of 6.3 years, 104 EC and 179 GC cases were observed, including 103 non-cardia gastric cancer (NCGC) cases and 76 cardia gastric cancer (CGC) cases. H. pylori-infected individuals exhibited a 1.78-fold increased risk of GC (HR 1.78, 95% confidence interval [CI] 1.32-2.40) but no significant increase in EC risk (HR 1.07, 95% CI 0.73-1.57). Notably, there was a higher risk for both NCGC and CGC in H. pylori-infected individuals. This population-based cohort study provides valuable evidence supporting the association between current H. pylori infection and the risk of both NCGC and CGC. These findings contribute to the empirical basis for risk stratification and recommendations for UGI cancer screening.

Bioimpedance based biomarker for the detection of precancerous and cancerous lesions of the pancreas: feasibility animal study

BackgroundPancreatic cancer (PC) remains a significant healthcare challenge due to its aggressive nature and poor prognosis. The current gold standard that combines imaging modalities, endoscopy, and biopsies has limited diagnostic efficacy due to various shortcomings.MethodsWe propose a feasibility study for the use of a bioimpedance biomarker to detect PC. The biomarker was evaluated in a double blind study on ex vivo pancreata of mice: 15 LSL-KrasG12D; LSL-p53R172H; Pdx1-Cre, 2 LSL-KrasG12D, and 9 wild type controls (Study 1). To determine if the biomarker can distinguish between PC and acute pancreatitis (AP), we challenged it with 18 cerulein-induced AP and 6 saline-injected controls (Study 2).ResultsThe results from Study 1 showed 100% specificity and 94% sensitivity against histopathology outcomes; for Study 2 all AP and saline-injected pancreases were diagnosed as non-cancerous. Regression analysis revealed a positive correlation between biomarker and pathologically analyzed cancer-induced fibrosis (r(24)= 0.73 (p<0.001)).ConclusionThese findings demonstrate the potential of this bioimpedance biomarker as a diagnostic tool for PC.

Abstract PO2-07-08: The efficacy of an MRI/US fusion technique for MRI-detected lesions previously undetected by conventional B-mode second-look US

Abstract Introduction Breast magnetic resonance imaging (MRI) is performed for preoperative screening of breast cancer and surveillance of BRCA gene carriers, and some lesions, called MRI-detected lesions, can only be detected by MRI. When MRI-detected lesions are found, second-look ultrasound (US) is recommended as the first choice for confirmation. However, the detection of MRI-detected lesions by conventional B-mode (cB-mode) second-look US varies from institution to institution, and it is often difficult to ensure objectivity and reproducibility. We use an MRI/US fusion technique called real-time virtual sonography (RVS) for the detection of MRI-detected lesions. RVS uses a magnetic position tracking system to display real-time US images on the same monitor as MRI images. Few reports on second-look US using RVS have examined its detection rate of MRI-detected lesions and the histopathological results of these lesions. In this study, we retrospectively reviewed MRI-detected lesions identified by second-look US using RVS but not by cB-mode second-look US from January 2018 to December 2022, with the goal of evaluating these lesions’ pathological characteristics. Materials and methods Consecutive patients who had one or more MRI-detected lesions not detected by cB-mode second-look US were enrolled in this study between January 2018 and December 2021. Second-look US using RVS was conducted after an additional supine MRI with a body surface coil was performed. Results A total of 36 patients with 38 lesions were included in this study. The mean age was 54 years old (range: 31–76). We were able to detect 31 (82%) of the 39 MRI-detected lesions by second-look US using RVS. These lesions were characterized as follows: mass, 14; non-mass-enhancement (NME), 13; and focal, 4. US-guided biopsy or excisional biopsy was performed on all lesions, and 11 (35%) of 31 lesions were malignant. We were able to detect 14 (74%) of the 19 mass lesions, and 3 were malignant (invasive ductal carcinoma (IDC) [luminal A type] in all 3 cases). We detected 13 (93%) of the 14 NME lesions, and 7 were malignant: 6 were ductal carcinoma in situ (DCIS) (high grade, 3; intermediate grade, 2; and low grade, 1), and 1 was IDC (luminal A type). Finally, we were able to detect 4 (80%) of the 5 focal lesions, 1 of which was malignant (DCIS [low grade]). Of 22 patients aged 55 years or younger, 3 had malignant lesions (IDC [luminal A type] in 2, DCIS [high grade] in 1). Of 15 patients aged 56 years or older, 8 had malignant lesions (DCIS in 7 [high grade, 2; intermediate grade, 2; low grade, 1] and IDC [luminal A type] in 1). Conclusions In this study, second-look US using RVS identified 82% of MRI-detected lesions that were not identified with cB-mode second-look US. Of these MRI-detected lesions, 35% were malignant. These results suggest that second-look US using RVS is useful for identifying MRI-detected lesions that are not detected by cB-mode second-look US. Citation Format: Masayuki Saito, Hirona Banno, Yukie Ito, Mirai Ido, Manami Goto, Takahito Ando, Yukako Mouri, Junko Kousaka, Kmihito Fujii, Tsuneo Imai, Shogo Nakano. The efficacy of an MRI/US fusion technique for MRI-detected lesions previously undetected by conventional B-mode second-look US [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-07-08.

Development and evaluation of RFID-integrated endoscopic clips for laparoscopic surgery marking.

As advancements in surgical instruments and techniques continue to evolve, minimally invasive surgery has become increasingly preferred as a means of reducing patient pain and recovery time. However, one major challenge in performing minimally invasive surgery for early gastrointestinal cancer is accurately identifying the location of the lesion. This is particularly difficult when the lesion is confined to the lumen of the intestine and cannot be visually confirmed from the outside during surgery. In such cases, surgeons must rely on CT or endoscopic imaging to locate the lesion. However, if the lesion is difficult to identify with these images or if the surgeon has less experience, it can be challenging to determine its precise location. This can result in an excessive resection margin, deviating from the goal of minimally invasive surgery. To address this challenge, researchers have been studying the development of a marker for identifying the lesion using a radio-frequency identification (RFID) system. One proposed method for clinical application of this detection system is to attach an RFID tag to an endoscopic hemostatic clip and fix it to the intended position, providing a stable marker for the inner wall of the organ. This approach has the potential to improve the accuracy and effectiveness of minimally invasive surgery for early gastrointestinal cancer. In the development of a marker for identifying gastrointestinal lesions using a radio-frequency identification (RFID) system, the shape of the clip and suitable materials for attaching the RFID tag were determined through finite element method (FEM) analysis. A prototype of the clip was then fabricated and ex-vivo experiments were conducted using porcine intestine to evaluate the stability of the clip in relation to its position. To further evaluate the performance of the RFID-integrated clip in vivo, the clip was placed in the gastric wall of the stomach of anesthetized porcine using an endoscopic instrument. The clip was then detected using a RFID detector designed for laparoscopic approach. And later, the accuracy of detection was confirmed by incising the lesion. The design and fabrication of a clip with varying thicknesses using STS316 and STS304 stainless steel were accomplished using the results of finite element method analysis. The stability of the clip was evaluated through ex-vivo experiments, showing it to be a viable option. In-vivo experiments were performed on anesthetized porcine, in which the RFID-integrated clip was placed in the gastric wall and detected using a custom-made RFID detector. The resection margin, measured at about 30 mm from the detector position, was accomplished with low error. These findings indicate the feasibility and efficacy of using an RFID-integrated clip as a marker in minimally invasive surgery for the identification of gastrointestinal lesions. The study evaluated the feasibility of using stainless steel clips for lesion detection in endoscopic surgery using computer-aided engineering analysis and ex-vivo experimentation. Results showed that STS304 was suitable for use while STS316L was not. The ex-vivo experiments revealed that the clip holding force and tissue retention length varied depending on the location of attachment. In-vivo experiments confirmed the accuracy and usefulness of the RFID lesion detection system. However, challenges remain for its use in clinical field, such as ensuring the stability of the clip and the safe attachment of the RFID tag, which requires further research for commercialization.

Abstract ED02-03: Innovative approaches at the diagnosis and treatment of invasive lobular carcinoma

Abstract Invasive Lobular Carcinoma (ILC) represents a unique subtype of breast cancer characterized by distinctive clinical and molecular features. This presentation aims to delve into innovative approaches in the diagnosis and treatment of ILC, focusing on novel techniques in imaging, unique molecular findings, signaling pathway alterations, targetable biomarkers, and microenvironment profiling that can pave the way for the development of novel therapies. Imaging Techniques in ILC: Accurate diagnosis is the foundation for effective treatment. ILC often presents diagnostic challenges due to the fact that lobular cells tends to infiltrate the affected organs in a diffuse manner instead of forming a discrete tumor mass making diagnostic imaging such as screening mammograms and CT scans less precise. This presentation will explore cutting-edge imaging techniques such as contrast-enhanced mammography, novel radioisotopes for positron emission tomography (PET) such as Fluoroestradiol F18 (FES), Fluciclovine and prostate-specific membrane antigen (PSMA) for improved detection and characterization of ILC lesions. Furthermore, we will discuss emerging modalities like radiomics and artificial intelligence, which hold promise in enhancing early diagnosis and risk assessment. Molecular Insights into ILC: ILC is distinguished by its unique molecular profile, notably the loss of E-cadherin expression. We will delve into recent advancements in understanding the molecular landscape of ILC, including genomic alterations, epigenetic changes, and the role of signaling pathways providing insights into potential therapeutic vulnerabilities. Targetable Biomarkers and Pathways: Identification of targetable biomarkers is essential for the development of precision therapies in ILC. This presentation will highlight promising targets, including ERBB2 (HER2), ROS1, TROP2, AKT1 along with other novel biomarkers. Microenvironment Profiling: The tumor microenvironment plays a pivotal role in cancer progression and therapy response. We will discuss recent advancements in characterizing the ILC microenvironment, including the immune cell infiltrate and stromal components. Understanding the dynamic interplay between the tumor and its microenvironment can inform the development of immunotherapies and combination strategies tailored to ILC. In conclusion, this presentation will provide a comprehensive overview of the latest innovations in ILC diagnosis and treatment. Our exploration of advanced imaging techniques, molecular findings, targetable pathways, biomarkers, and microenvironment profiling will shed light on the exciting potential for personalized therapeutic approaches in the management of ILC. By staying at the forefront of these developments, we can pave the way for improved outcomes and enhanced quality of life for patients with ILC. Citation Format: J. Mouabbi. Innovative approaches at the diagnosis and treatment of invasive lobular carcinoma [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr ED02-03.

Past, Present, and Future Diagnostic Methods for the Early Noninvasive Detection of Oral Premalignant Lesions: A State of the Art and Systematic Review.

Objectives: To provide an in-depth analysis of noninvasive methods for the early diagnosis of oral premalignant lesions, focusing on novel biomarkers and optical technologies, and to discuss their potential in improving the prognosis of patients with oral oncological diseases. Methods: This state-of-the-art review examines various noninvasive diagnostic techniques, including the utilization of salivary microRNAs and optical technologies such as Raman spectroscopy, elastic scattering spectroscopy, diffuse reflectance spectroscopy, narrow-band imaging, autofluorescence imaging, toluidine blue staining, and microendoscopy. Results: Several noninvasive techniques have shown varying degrees of effectiveness in detecting oral cancer. Autofluorescence imaging exhibited sensitivities up to 100% but had variable specificity. toluidine blue staining reported sensitivity between 77% and 100% for high-risk lesions or cancer, with specificity around 45% to 67%. Spectroscopy techniques achieved 72% to 100% sensitivities and specificities of 75% to 98%. Microendoscopy presented a sensitivity of 84% to 95% and a specificity of 91% to 95%. Conclusion: The review highlights the strengths and limitations of each noninvasive diagnostic method and their recent advancements. Although promising results have been demonstrated, there is a need for further development of reliable strategies for early detection and intervention in oral oncology.

Abstract PO5-19-11: A novel portable breast cancer screening method - initial institutional experience

Abstract Background: Breast ultrasound (US) is a remarkable supplementary tool alongside mammography and magnetic resonance imaging (MRI) for the detection of breast lesions. US imaging is a fast, reliable method to identify and localize breast pathology. Innovative technologies that aim to timely identification and prompt diagnosis of suspicious lesions in breast cancer screening remain a clinical need. ATUSA is a B-type rechargeable US, shaped as a wearable device. It uses a linear array of transducers with water acting as a coupling medium that recreates a 3-dimensional image of the scanned breast. The ATUSA portability makes this device a practical breast screening method during a healthcare office visit. We report the initial feasibility of ATUSA to detect breast lesions in an outpatient setting from a trial in progress. Methodology: Interim prospective data from 35 female patients presenting with a mass of equal or greater than 1 cm in size, with a baseline handheld US image and pathology reports that evidenced a cancer diagnosis or a lesion suspicious of malignancy. Breast cancer types included invasive ductal carcinoma (IDC), ductal carcinoma in situ (DCIS) and invasive lobular carcinoma (ILC), invasive mammary carcinoma (IMC) with mixed ductal and lobular features and atypical lobular hyperplasia (ALH). ATUSA and baseline handheld reference US images were cross-referenced from June 2022 to June 2023. Basic demographics, clinical characteristics, mammogram and/or MRI, and pathology reports were included in the analysis. Patients were followed up for a 30-day period to note occurrence of any adverse event (AE). Institutional Review Board (IRB) approved this study (protocol H-42616). Results: The 35 patients had a mean age of 57 years (range: 25 – 79), mean BMI 29.6 kg/m2 (range: 18.1 – 49.3). Mammographic breast density was fibroglandular tissue 15 (42.8%), followed by heterogeneously dense (HD) 13 (3.7%), mixed HD and fibroglandular 1 (2.8%), mildly dense 1 (2.8%), extremely dense 2 (5.7%), and extremely fatty 1 (2.8%). Breast sizes were 1 (2.8 %) A cup, 11 (31.4%) B cups, 1 (2.8%) B – C cups, 5 (14.2%) C cups, 1 (2.8%) C – D cups, 13 (37.1%) D cups, and 3 (8.5%) DD cups. The 35 scanned lesions had a diagnosis (compared with final resected pathology) classified as IDC 18 (51.4 %), DCIS 2 (5.7 %), mixed IDC and DCIS 6 (17.1 %), ILC 3 (8.5 %), IMC with mixed features 2 (5.7 %), IMC with DCIS 1 (2.8 %), ALH 1 (2.8 %), and 2 (5.7 %) benign masses. ATUSA successfully identified 17/35 (48.5 %) lesions from this group as malignant, from which 4 were hard to distinguish due to shadowing artifact and 3 were underestimated in size. A total of 15 (42.8 %) malignant lesions were missed by ATUSA. No AEs related to the ATUSA device were reported. Conclusion: The ATUSA provides the means to produce a good quality US image during a clinic visit or where a facility lacks conventional US imaging. When compared to the handheld US, ATUSA can be safely used in the outpatient setting without the need of trained specialists but requires further development and testing to match conventional breast imaging. Table 1. Basic clinical characteristics &amp; demographics Citation Format: Margarita Riojas-Barrett, Jessica Montalvan, Mary Bajomo, Ivan Marin, Quan Dang Nguyen, Karla A Sepulveda, Elizabeth Bonefas, Alastair Thompson, Stacey Carter. A novel portable breast cancer screening method - initial institutional experience [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-19-11.

Abstract PS10-06: Contribution of a breast cancer polygenic risk score to mammography artificial intelligence models and breast density for long term breast cancer risk prediction

Abstract Artificial intelligence (AI) models based on deep learning of mammogram images have been developed for lesion detection and diagnosis as well as breast cancer (BC) risk. We and others have shown that these AI models, coupled with breast density, predict long-term breast cancer risk. Common germline genetic variation, in particular a polygenic risk score (PRS) that captures the cumulative impact of hundreds of common risk variants, has been shown associated with 3-fold increased BC risk. The BC-PRS may complement imaging risk factors to improve identification of women at high BC risk and better target prevention strategies. We evaluated the contribution of a BC-PRS to two imaging-based AI models (Transpara AI cancer detection system and Mirai 5-year risk) and breast density measures to long-term BC risk. We hypothesized that the BC-PRS will improve long-term risk prediction above AI and density measures. We examined this hypothesis within a cohort of 10,271 women enrolled to the Mayo Clinic Biobank between 2009 and 2015. Eligible women had no prior cancer, were ages 35 to 85, had a full-field digital screening mammogram within two years of enrollment and genotyping by Regeneron. Incident invasive and DCIS BC was identified through the Mayo tumor registry and follow-up questionnaires, and person years were calculated from date of mammogram through date of BC, last mammogram at Mayo or last questionnaire. We estimated Transpara AI malignancy score (1-10), 5-year Mirai risk and Volpara volumetric density measures (volumetric percent density, dense volume) on screening mammograms at enrollment and obtained clinical BI-RADS density. We calculated the BC-PRS using 268 SNPs (of the 313-SNP score) and the weighted sum of the SNPs using odds ratios from the largest BC GWAS of European ancestry to date. We performed cox proportional hazards regression adjusting for age and BMI to estimate hazard ratios (HRs), 95% confidence intervals (CI), and C-statistics (AUC) to describe the contribution of BC-PRS (per standard deviation (SD)) to the association of AI scores and breast density with BC risk. Likelihood ratio tests (LRT) and bootstrapping were used to compare model performance with vs. without inclusion of BC-PRS, at 10-years follow-up. Comparisons to the Tyrer-Cuzick model are ongoing. Over a median of 9.1 years (Interquartile range, 6.8 to 10.6), 446 incident BC were identified of which 285 invasive and 103 DCIS were confirmed by pathology report (60 were self-reported); 399 (256 invasive) occurred within 10 years and were used in primary analyses. Transpara score [HR per 1 unit=1.18 (1.14, 1.23), AUC=0.655 (0.628, 0.682)] and Mirai risk [HR per SD log risk=1.54 (1.43, 1.65), AUC=0.681 (0.655, 0.706)] were both associated with BC. All three breast density measures were also significantly associated with BC, including in models with AI scores (Table). Addition of the BC-PRS to models with AI scores and breast density improved both model fit (All PLRT &amp;lt; 0.001; Table) and discrimination, with increases in AUC ranging from 0.033 to 0.034 for models with Transpara and density and 0.025 to 0.026 for models with Mirai and density (all P-values &amp;lt; 0.001; Table). Analyses restricted to invasive BC showed identical results. But, the BC-PRS contribution at 5-year follow-up was attenuated. In this cohort study, we found BC-PRS contributed to long-term risk prediction of BC above imaging AI models and breast density. Imaging and genetic risk factors are complementary and may outperform existing risk models and provide more accurate risk assessment to aid in precision prevention strategies. Citation Format: Celine Vachon, Christopher Scott, Imon Banerjee, Ramon Correa Moderno, Dan Hursh, Matthew Jensen, Yvonne Wang, Kathleen Brandt, Karla Kerlikowske, Sandhya Pruthi, Fergus Couch, Stacey Winham. Contribution of a breast cancer polygenic risk score to mammography artificial intelligence models and breast density for long term breast cancer risk prediction [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS10-06.

Evaluation of a Cascaded Deep Learning-based Algorithm for Prostate Lesion Detection at Biparametric MRI.

Background Multiparametric MRI (mpMRI) improves prostate cancer (PCa) detection compared with systematic biopsy, but its interpretation is prone to interreader variation, which results in performance inconsistency. Artificial intelligence (AI) models can assist in mpMRI interpretation, but large training data sets and extensive model testing are required. Purpose To evaluate a biparametric MRI AI algorithm for intraprostatic lesion detection and segmentation and to compare its performance with radiologist readings and biopsy results. Materials and Methods This secondary analysis of a prospective registry included consecutive patients with suspected or known PCa who underwent mpMRI, US-guided systematic biopsy, or combined systematic and MRI/US fusion-guided biopsy between April 2019 and September 2022. All lesions were prospectively evaluated using Prostate Imaging Reporting and Data System version 2.1. The lesion- and participant-level performance of a previously developed cascaded deep learning algorithm was compared with histopathologic outcomes and radiologist readings using sensitivity, positive predictive value (PPV), and Dice similarity coefficient (DSC). Results A total of 658 male participants (median age, 67 years [IQR, 61-71 years]) with 1029 MRI-visible lesions were included. At histopathologic analysis, 45% (294 of 658) of participants had lesions of International Society of Urological Pathology (ISUP) grade group (GG) 2 or higher. The algorithm identified 96% (282 of 294; 95% CI: 94%, 98%) of all participants with clinically significant PCa, whereas the radiologist identified 98% (287 of 294; 95% CI: 96%, 99%; P = .23). The algorithm identified 84% (103 of 122), 96% (152 of 159), 96% (47 of 49), 95% (38 of 40), and 98% (45 of 46) of participants with ISUP GG 1, 2, 3, 4, and 5 lesions, respectively. In the lesion-level analysis using radiologist ground truth, the detection sensitivity was 55% (569 of 1029; 95% CI: 52%, 58%), and the PPV was 57% (535 of 934; 95% CI: 54%, 61%). The mean number of false-positive lesions per participant was 0.61 (range, 0-3). The lesion segmentation DSC was 0.29. Conclusion The AI algorithm detected cancer-suspicious lesions on biparametric MRI scans with a performance comparable to that of an experienced radiologist. Moreover, the algorithm reliably predicted clinically significant lesions at histopathologic examination. ClinicalTrials.gov Identifier: NCT03354416 © RSNA, 2024 Supplemental material is available for this article.

Leveraging a realistic synthetic database to learn Shape-from-Shading for estimating the colon depth in colonoscopy images

Colonoscopy is the choice procedure to diagnose, screening, and treat the colon and rectum cancer, from early detection of small precancerous lesions (polyps), to confirmation of malign masses. However, the high variability of the organ appearance and the complex shape of both the colon wall and structures of interest make this exploration difficult. Learned visuospatial and perceptual abilities mitigate technical limitations in clinical practice by proper estimation of the intestinal depth. This work introduces a novel methodology to estimate colon depth maps in single frames from monocular colonoscopy videos. The generated depth map is inferred from the shading variation of the colon wall with respect to the light source, as learned from a realistic synthetic database. Briefly, a classic convolutional neural network architecture is trained from scratch to estimate the depth map, improving sharp depth estimations in haustral folds and polyps by a custom loss function that minimizes the estimation error in edges and curvatures. The network was trained by a custom synthetic colonoscopy database herein constructed and released, composed of 248400 frames (47 videos), with depth annotations at the level of pixels. This collection comprehends 5 subsets of videos with progressively higher levels of visual complexity. Evaluation of the depth estimation with the synthetic database reached a threshold accuracy of 95.65%, and a mean-RMSE of 0.451cm, while a qualitative assessment with a real database showed consistent depth estimations, visually evaluated by the expert gastroenterologist coauthoring this paper. Finally, the method achieved competitive performance with respect to another state-of-the-art method using a public synthetic database and comparable results in a set of images with other five state-of-the-art methods. Additionally, three-dimensional reconstructions demonstrated useful approximations of the gastrointestinal tract geometry. Code for reproducing the reported results and the dataset are available at https://github.com/Cimalab-unal/ColonDepthEstimation.

A narrative review on the current uses of artificial intelligence in endodontics

Abstract Artificial intelligence (AI) has been widely introduced to dentistry in the past decade. Its application in endodontics is limited to different areas such as working length determination, morphological assessment, detection of vertical root fracture, and the detection of periapical lesion. Therefore, this study aims to highlight the available evidence for the uses of AI in endodontics. It also presents the current status as well as the future perspectives on the uses of AI and its potential application in everyday practice. A literature search was conducted from January 2000 to January 2023 using PubMed and Google Scholar for the terms AI and endodontics. Thirty-one studies were evaluated and summarized, highlighting the potential use of different AI models in endodontics. The evaluation of the studies indicated that the use of AI is promising and could aid in tailored endodontics therapy. It would help the clinician in the detection of periapical radiolucency, root fractures, and determination of working length. However, well-designed, high-quality research is required to assess the possible implementation of AI into day-to-day practice in endodontics.

Multiscale and multiperception feature learning for pancreatic lesion detection based on noncontrast CT

Background. Pancreatic cancer is one of the most malignant tumours, demonstrating a poor prognosis and nearly identically high mortality and morbidity, mainly because of the difficulty of early diagnosis and timely treatment for localized stages. Objective. To develop a noncontrast CT (NCCT)-based pancreatic lesion detection model that could serve as an intelligent tool for diagnosing pancreatic cancer early, overcoming the challenges associated with low contrast intensities and complex anatomical structures present in NCCT images. Approach. We design a multiscale and multiperception (MSMP) feature learning network with ResNet50 coupled with a feature pyramid network as the backbone for strengthening feature expressions. We added multiscale atrous convolutions to expand different receptive fields, contextual attention to perceive contextual information, and channel and spatial attention to focus on important channels and spatial regions, respectively. The MSMP network then acts as a feature extractor for proposing an NCCT-based pancreatic lesion detection model with image patches covering the pancreas as its input; Faster R-CNN is employed as the detection method for accurately detecting pancreatic lesions. Main results. By using the new MSMP network as a feature extractor, our model outperforms the conventional object detection algorithms in terms of the recall (75.40% and 90.95%), precision (40.84% and 68.21%), F1 score (52.98% and 77.96%), F2 score (64.48% and 85.26%) and Ap50 metrics (53.53% and 70.14%) at the image and patient levels, respectively. Significance.The good performance of our new model implies that MSMP can mine NCCT imaging features for detecting pancreatic lesions from complex backgrounds well. The proposed detection model is expected to be further developed as an intelligent method for the early detection of pancreatic cancer.

3DFRINet: A Framework for the Detection and Diagnosis of Fracture Related Infection in Low Extremities Based on 18F-FDG PET/CT 3D Images

Fracture related infection (FRI) is one of the most devastating complications after fracture surgery in the lower extremities, which can lead to extremely high morbidity and medical costs. Therefore, early comprehensive evaluation and accurate diagnosis of patients are critical for appropriate treatment, prevention of complications, and good prognosis. 18Fluoro-deoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) is one of the most commonly used medical imaging modalities for diagnosing FRI. With the development of deep learning, more neural networks have been proposed and become powerful computer-aided diagnosis tools in medical imaging. Therefore, a fully automated two-stage framework for FRI detection and diagnosis, 3DFRINet (Three Dimension FRI Network), is proposed for 18F-FDG PET/CT 3D imaging. The first stage can effectively extract and fuse the features of both modalities to accurately locate the lesion by the dual-branch design and attention module. The second stage reduces the dimensionality of the image by using the maximum intensity projection, which retains the effective features while reducing the computational effort and achieving excellent diagnostic performance. The diagnostic performance of lesions reached 91.55% accuracy, 0.9331 AUC, and 0.9250 F1 score. 3DFRINet has an advantage over six nuclear medicine experts in each classification metric. The statistical analysis shows that 3DFRINet is equivalent or superior to the primary nuclear medicine physicians and comparable to the senior nuclear medicine physicians. In conclusion, this study first proposed a method based on 18F-FDG PET/CT three-dimensional imaging for FRI location and diagnosis. This method shows superior lesion detection rate and diagnostic efficiency and therefore has good prospects for clinical application.

YOLOv8-RMDA: Lightweight YOLOv8 Network for Early Detection of Small Target Diseases in Tea

In order to efficiently identify early tea diseases, an improved YOLOv8 lesion detection method is proposed to address the challenges posed by the complex background of tea diseases, difficulty in detecting small lesions, and low recognition rate of similar phenotypic symptoms. This method focuses on detecting tea leaf blight, tea white spot, tea sooty leaf disease, and tea ring spot as the research objects. This paper presents an enhancement to the YOLOv8 network framework by introducing the Receptive Field Concentration-Based Attention Module (RFCBAM) into the backbone network to replace C2f, thereby improving feature extraction capabilities. Additionally, a mixed pooling module (Mixed Pooling SPPF, MixSPPF) is proposed to enhance information blending between features at different levels. In the neck network, the RepGFPN module replaces the C2f module to further enhance feature extraction. The Dynamic Head module is embedded in the detection head part, applying multiple attention mechanisms to improve multi-scale spatial location and multi-task perception capabilities. The inner-IoU loss function is used to replace the original CIoU, improving learning ability for small lesion samples. Furthermore, the AKConv block replaces the traditional convolution Conv block to allow for the arbitrary sampling of targets of various sizes, reducing model parameters and enhancing disease detection. the experimental results using a self-built dataset demonstrate that the enhanced YOLOv8-RMDA exhibits superior detection capabilities in detecting small target disease areas, achieving an average accuracy of 93.04% in identifying early tea lesions. When compared to Faster R-CNN, MobileNetV2, and SSD, the average precision rates of YOLOv5, YOLOv7, and YOLOv8 have shown improvements of 20.41%, 17.92%, 12.18%, 12.18%, 10.85%, 7.32%, and 5.97%, respectively. Additionally, the recall rate (R) has increased by 15.25% compared to the lowest-performing Faster R-CNN model and by 8.15% compared to the top-performing YOLOv8 model. With an FPS of 132, YOLOv8-RMDA meets the requirements for real-time detection, enabling the swift and accurate identification of early tea diseases. This advancement presents a valuable approach for enhancing the ecological tea industry in Yunnan, ensuring its healthy development.

Determination of the timing for thoracic imaging prior to pulmonary metastasectomy: an analysis on surgical planning and lesion detection

Aims: The aim of this study is to reveal the relationship between the timing of thoracic computed tomography (CT) imaging conducted prior to surgery and the pre-surgical period in patients planned for pulmonary metastasectomy (PM), and to determine a safe pre-surgical timing for thoracic CT. Methods: This study is a retrospective cohort study examining the data of patients who underwent pulmonary metastasectomy (PM). The research includes 96 patients who underwent PM between January 2017 and July 2022. Patients’ demographic data, primary malignancy diagnoses, type of operation, sizes of masses requiring anatomical resection, the number of lesions detected in thoracic CT, the number of lesions identified during surgery, and the timing of thoracic tomography were recorded. The timing of thoracic CT imaging was compared with the number of lesions detected preoperatively and postoperatively. Results: The study included 96 patients, comprising 49 females and 47 males. The most common primary pathological diagnosis was colon cancer at 36.5%, followed by breast cancer at 12.5%. 66.6% of the patients were operated on with thoracotomy, 29.1% with video-assisted thoracoscopic surgery (VATS), and 4.2% with rethoracotomy. The average number of lesions detected in preoperative thoracic tomography was 1.67±0.96, while the average number of lesions detected during surgery was 2.03±1.41. In patient groups where thoracic CT was performed 10 days or less before the operation, no significant difference was found between the number of lesions detected during surgery and the number of lesions in the CT. However, in patients where thoracic CT was performed more than 10 days before the operation, the number of lesions detected during surgery was significantly higher than the number of lesions detected in the CT. Conclusion: In this research, it was concluded that for patients planned for PM, repeating thoracic CT after the 10th day following the initial detection of metastases in the pre-surgical phase may contribute to the detection of more lesions.

An enhanced skin lesion detection and classification model using hybrid convolution-based ensemble learning model

An enhanced skin lesion detection and classification model using hybrid convolution-based ensemble learning model

Awareness and Practice of Breast Self-examination for Early Detection of Breast Lesions in Pregnant Women

Awareness and Practice of Breast Self-examination for Early Detection of Breast Lesions in Pregnant Women

Detailed Superiority of the CAD EYE Artificial Intelligence System over Endoscopists for Lesion Detection and Characterization Using Unique Movie Sets.

Detailed superiority of CAD EYE (Fujifilm, Tokyo, Japan), an artificial intelligence for polyp detection/diagnosis, compared to endoscopists is not well examined. We examined endoscopist's ability using movie sets of colorectal lesions which were detected and diagnosed by CAD EYE accurately. Consecutive lesions of ≤10 mm were examined live by CAD EYE from March-June 2022 in our institution. Short unique movie sets of each lesion with and without CAD EYE were recorded simultaneously using two recorders for detection under white light imaging (WLI) and linked color imaging (LCI) and diagnosis under blue laser/light imaging (BLI). Excluding inappropriate movies, 100 lesions detected and diagnosed with CAD EYE accurately were evaluated. Movies without CAD EYE were evaluated first by three trainees and three experts. Subsequently, movies with CAD EYE were examined. The rates of accurate detection and diagnosis were evaluated for both movie sets. Among 100 lesions (mean size: 4.7±2.6 mm; 67 neoplastic/33 hyperplastic), mean accurate detection rates of movies without or with CAD EYE were 78.7%/96.7% under WLI (p<0.01) and 91.3%/97.3% under LCI (p<0.01) for trainees and 85.3%/99.0% under WLI (p<0.01) and 92.6%/99.3% under LCI (p<0.01) for experts. Mean accurate diagnosis rates of movies without or with CAD EYE for BLI were 85.3%/100% for trainees (p<0.01) and 92.3%/100% for experts (p<0.01), respectively. The significant risk factors of not-detected lesions for trainees were right-sided, hyperplastic, not-reddish, in the corner, halation, and inadequate bowel preparation. Unique movie sets with and without CAD EYE could suggest it's efficacy for lesion detection/diagnosis.

Effect of 18F-DCFPyL PET on changes in management of patients with prostate cancer: a systematic review and meta-analysis

PurposeProstate-specific membrane antigen (PSMA)-targeted imaging has gained increasing interest in its application in prostate cancer lesion detection. Compared with 68Galium (68Ga), 18Fluoride (18F)-labeled imaging agent has easier syntheses, lower price, and a longer half-time. 2-(3-{1-Carboxy-5-[(6-[18F]fluoro-pyridine-3-carbonyl)-amino]-pentyl}-ureido)-pentanedioic acid positron emission tomography (18F-DCFPyL PET) has been recently approved by the U.S. Food and Drug Administration. Several studies have proven its superiority to conventional imaging techniques in detecting prostate cancer lesions. However, the impact of 18F-DCFPyL PET on the management of patients with prostate cancer is not well established. Thus, we performed a systematic review and meta-analysis of available data to evaluate the impact of 18F-DCFPyL PET on the management of patients with prostate cancer.MethodsThe PubMed, Embase, Scopus, and Cochrane databases were searched up to April 2024. Studies that reported the proportion of changes in management after 18F-DCFPyL PET was performed in patients with prostate cancer were included. The Grading of Recommendations Assessment, Development, and Evaluation system was used for the quality evaluation of the included studies. The proportion of changes in management was pooled using a random effects model. Meta-regression analyses were performed to assess the potential correlation between the PET positivity and management changes.ResultsFourteen studies (3,078 patients with prostate cancer) were included in our review and analysis. The pooled percentage of management changes was 43.5% (95% confidence interval [CI]: 33–54%). In patients with biochemical recurrent and for primary staging, the pooled percentage was 50% (95% CI: 39–60%) and 22% (95% CI: 15–29%), respectively. In the meta-regression analyses, PET positivity was detected as a significant predictor of management change (p = 0.0023).Conclusion18F-DCFPyL PET significantly affects the management of patients with prostate cancer. Higher PET positivity rate significantly correlated with a higher proportion of management changes in patients with prostate cancer. However, more studies are still needed to confirm the important role of 18F-DCFPyL PET in the management of prostate cancer.Systematic review registrationhttps://www.crd.york.ac.uk/PROSPERO/#myprospero, CRD42022339178.

Patlak Slope versus Standardized Uptake Value Image Quality in an Oncologic PET/CT Population: A Prospective Cross-Sectional Study

Patlak slope (PS) images have the potential to improve lesion conspicuity compared with standardized uptake value (SUV) images but may be more artifact-prone. This study compared PS versus SUV image quality and hepatic tumor-to-background ratios (TBRs) at matched time points. Early and late SUV and PS images were reconstructed from dynamic positron emission tomography (PET) data. Two independent, blinded readers scored image quality metrics (a four-point Likert scale) and counted tracer-avid lesions. Hepatic lesions and parenchyma were segmented and quantitatively analyzed. Differences were assessed via the Wilcoxon signed-rank test (alpha, 0.05). Forty-three subjects were included. For overall quality and lesion detection, early PS images were significantly inferior to other reconstructions. For overall quality, late PS images (reader 1 [R1]: 3.95, reader 2 [R2]: 3.95) were similar (p &gt; 0.05) to early SUV images (R1: 3.88, R2: 3.84) but slightly superior (p ≤ 0.002) to late SUV images (R1: 2.97, R2: 3.44). For lesion detection, late PS images were slightly inferior to late SUV images (R1 only) but slightly superior to early SUV images (both readers). PS-based TBRs were significantly higher than SUV-based TBRs at the early time point, with opposite findings at the late time point. In conclusion, late PS images are similar to early/late SUV images in image quality and lesion detection; the superiority of SUV versus PS hepatic TBRs is time-dependent.