Detection Of Lesions Research Articles

Evaluating the role of 7-Tesla magnetic resonance imaging in neurosurgery: Trends in literature since clinical approval.

After approval for clinical use in 2017, early investigations of ultra-high-field abdominal magnetic resonance imaging (MRI) have demonstrated its feasibility as well as diagnostic capabilities in neuroimaging. However, there are no to few systematic reviews covering the entirety of its neurosurgical applications as well as the trends in the literature with regard to the aforementioned application. To assess the impact of 7-Tesla MRI (7T MRI) on neurosurgery, focusing on its applications in diagnosis, treatment planning, and postoperative assessment, and to systematically analyze and identify patterns and trends in the existing literature related to the utilization of 7T MRI in neurosurgical contexts. A systematic search of PubMed was conducted for studies published between January 1, 2017, and December 31, 2023, using MeSH terms related to 7T MRI and neurosurgery. The inclusion criteria were: Studies involving patients of all ages, meta-analyses, systematic reviews, and original research. The exclusion criteria were: Pre-prints, studies with insufficient data (e.g., case reports and letters), non-English publications, and studies involving animal subjects. Data synthesis involved standardized extraction forms, and a narrative synthesis was performed. We identified 219 records from PubMed within our defined period, with no duplicates or exclusions before screening. After screening, 125 articles were excluded for not meeting inclusion criteria, leaving 94 reports. Of these, 2 were irrelevant to neurosurgery and 7 were animal studies, resulting in 85 studies included in our systematic review. Data were categorized by neurosurgical procedures and diseases treated using 7T MRI. We also analyzed publications by country and the number of 7T MRI facilities per country was also presented. Experimental studies were classified into comparison and non-comparison studies based on whether 7T MRI was compared to lower field strengths. 7T MRI holds great potential in improving the characterization and understanding of various neurological and psychiatric conditions that may be neurosurgically treated. These include epilepsy, pituitary adenoma, Parkinson's disease, cerebrovascular diseases, trigeminal neuralgia, traumatic head injury, multiple sclerosis, glioma, and psychiatric disorders. Superiority of 7T MRI over lower field strengths was demonstrated in terms of image quality, lesion detection, and tissue characterization. Findings suggest the need for accelerated global distribution of 7T magnetic resonance systems and increased training for radiologists to ensure safe and effective integration into routine clinical practice.

Diagnostic efficacy of SEPT9 and PAX5 gene methylation in gastrointestinal cancer and precancerous lesions.

To assess the diagnostic efficacy of SEPT9 along with PAX5 gene methylation detection in gastrointestinal cancer and precancerous lesions, the peripheral blood of 62 patients with gastric cancer (GC) and 60 patients with no evidence of disease (as the control group) were retrospectively collected. The methylation rates of PAX5 and SEPT9 gene promoters in blood samples of GC group were detected by PCR. At the same time, the differences in methylation rates of genes in the two groups were compared, and the predictive value of plasma methylation PAX5 and SEPT9 in GC was evaluated by receiver operating characteristic (ROC) curve. We found that there were 41 cases of methylated PAX5 gene promoter region and 39 cases of methylated SEPT9 gene promoter region in GC group. The control group contained 14 cases of PAX5 gene promoter methylation and 12 cases of RNF¹80 gene promoter methylation. The occurrence of PAX5 promoter methylation was correlated with age of GC patients. There were statistically significant differences in mSEPT9 gene in patients with different TNM stages. Kaplan-Meier survival curve analysis revealed that the three-year overall survival rate of GC patients with PAX5 methylation was lower than that of GC patients without PAX5 methylation. No significant difference was discovered in 3-year overall survival rate between GC patients with SEPT9 methylation and those without SEPT9 methylation. Combined detection could not improve the diagnostic value of GC, but could promote diagnosis sensitivity. In summary, the risk of PAX5 and SEPT9 gene methylation in GC patients presents higher when compared with healthy people. PAX5 gene methylation is closely related to age, while SEPT9 is closely related to tumor TNM stage, and PAX5 gene methylation can decrease the survival rate of GC patients. Detection of PAX5 gene methylation level can assist in evaluating the prognosis of GC patients.

Expanding management strategies for cervical precancerous lesions in resource-limited settings: insights from a training center in a district hospital in Ghana

BackgroundCervical cancer continues to disproportionately burden women in low/middle-income countries like Ghana. We examined treatment patterns and histopathological outcomes among women screened using visual inspection with acetic acid (VIA) and/or mobile colposcopy who subsequently underwent thermal ablation, large loop excision of the transformation zone (LLETZ), or cold knife conization at the Cervical Cancer Prevention and Training Centre, Battor. We also assessed the prevalence of cervical intraepithelial neoplasia 2+ (CIN2+) or micro-invasive disease and their associated factors for women who underwent excisional treatments. The treatment choices for cervical precancerous lesions suitable for resource-limited settings have also been described from the perspective of a center that manages a heterogenous population.MethodsWe conducted an analysis of secondary data collected between June 2016 and June 2023 among women with positive findings on VIA or mobile colposcopy who subsequently underwent thermal ablation or large loop excision of the transformation zone (LLETZ). The prevalence of histopathology outcomes, including no dysplasia, CIN1 − 3, and micro-invasive disease, were estimated with 95% confidence intervals (CIs). Factors associated with histopathological findings were modeled using multinomial logistic regression.ResultsFor the study period, 14 (10.6%) of the total 132 participants underwent cervical lesion treatment at outreach locations, all via thermal ablation. The remaining 118 (89.4%) were treated at the Catholic Hospital, Battor using LLETZ (n = 66, 55.9%), thermal ablation (n = 51, 43.2%), and cold knife conization (n = 1, 0.9%). Among 65 women with histopathology reports, the most frequent histopathological finding was no dysplasia (47.7%; 95% CI, 35.1 − 60.5), followed by CIN2 and CIN3 (20.0%; 95% CI, 11.1 − 31.8 each), CIN1 (7.7%; 95% CI, 2.5 − 17.0) and micro-invasion (4.6%; 95% CI, 1.0 − 12.9). Those with micro-invasive disease were significantly older than those with CIN1, CIN2, and CIN3 (p = 0.036, 0.022, 0.009, respectively), but not significantly older than those who showed no dysplasia (p = 0.088). For each unit increase in age, the likelihood of CIN3 was relatively significantly reduced compared to no dysplasia (crude relative risk ratio [RRR] = 0.93; 95% CI, 0.86 − 0.99). This association was neither observed with the remaining histopathological groups nor for parity and persisted after controlling for parity (adjusted RRR = 0.92; 95% CI, 0.85 − 0.99; p = 0.025).ConclusionThis paper largely demonstrates treatment options available to women and practitioners in LMICs. The high combined prevalence of high-grade precancerous lesions and micro-invasive disease underscores the need to increase cervical cancer awareness that would enhance screening attendance and hasten efforts at moving from opportunistic to organized screening in Ghana. This will enhance early cervical lesion detection and treatment, while simultaneously re-evaluating and cutting down on unnecessary treatment.

Endoscopic Artificial Intelligence for Image Analysis in Gastrointestinal Neoplasms.

Artificial intelligence (AI) using deep learning systems has recently been utilized in various medical fields. In the field of gastroenterology, AI is primarily implemented in image recognition and utilized in the realm of gastrointestinal (GI) endoscopy. In GI endoscopy, computer-aided detection/diagnosis (CAD) systems assist endoscopists in GI neoplasm detection or differentiation of cancerous or non-cancerous lesions. Several AI systems for colorectal polyps have already been applied in colonoscopy clinical practices. In esophagogastroduodenoscopy, a few CAD systems for upper GI neoplasms have been launched in Asian countries. The usefulness of these CAD systems in GI endoscopy has been gradually elucidated. In this review, we outline recent articles on several studies of endoscopic AI systems for GI neoplasms, focusing on esophageal squamous cell carcinoma (ESCC), esophageal adenocarcinoma (EAC), gastric cancer (GC), and colorectal polyps. In ESCC and EAC, computer-aided detection (CADe) systems were mainly developed, and a recent meta-analysis study showed sensitivities of 91.2% and 93.1% and specificities of 80% and 86.9%, respectively. In GC, a recent meta-analysis study on CADe systems demonstrated that their sensitivity and specificity were as high as 90%. A randomized controlled trial (RCT) also showed that the use of the CADe system reduced the miss rate. Regarding computer-aided diagnosis (CADx) systems for GC, although RCTs have not yet been conducted, most studies have demonstrated expert-level performance. In colorectal polyps, multiple RCTs have shown the usefulness of the CADe system for improving the polyp detection rate, and several CADx systems have been shown to have high accuracy in colorectal polyp differentiation. Most analyses of endoscopic AI systems suggested that their performance was better than that of non-expert endoscopists and equivalent to that of expert endoscopists. Thus, Endoscopic AI systems may be useful for reducing the risk of overlooking lesions and improving the diagnostic ability of endoscopists.

Deep learning-assisted multispectral imaging for early screening of skin diseases

IntroductionMelanocytic nevi (MN), warts, seborrheic keratoses (SK), and psoriasis are four common types of skin surface lesions that typically require dermatoscopic examination for definitive diagnosis in clinical dermatology settings. This process is labor-intensive and resource-consuming. Traditional methods for diagnosing skin lesions rely heavily on the subjective judgment of dermatologists, leading to issues in diagnostic accuracy and prolonged detection times. ObjectivesThis study aims to introduce a multispectral imaging (MSI)-based method for the early screening and detection of skin surface lesions. By capturing image data at multiple wavelengths, MSI can detect subtle spectral variations in tissues, significantly enhancing the differentiation of various skin conditions. MethodsThe proposed method utilizes a pixel-level mosaic imaging spectrometer to capture multispectral images of lesions, followed by reflectance calibration and standardization. Regions of interest were manually extracted, and the spectral data were subsequently exported for analysis. An improved one-dimensional convolutional neural network is then employed to train and classify the data. ResultsThe new method achieves an accuracy of 96.82 % on the test set, demonstrating its efficacy. ConclusionThis multispectral imaging approach provides a non-contact and non-invasive method for early screening, effectively addressing the subjective identification of lesions by dermatologists and the prolonged detection times associated with conventional methods. It offers enhanced diagnostic accuracy for a variety of skin lesions, suggesting new avenues for dermatological diagnostics.

SISCOS in focal cortical dysplasia: localization and comparative analysis with MRI.

This study evaluates the efficacy of SISCOS (Subtraction ictal-interictal SPECT coregistered to SPECT) in localizing the epileptogenic zone (EZ) in focal cortical dysplasia (FCD), comparing its predictive performance with MRI and post-surgical outcomes based on ILAE classification. 84 patients with drug refractory epilepsy (DRE) who were operated and had histopathology consistent with FCD, were included in the study. All patients had undergone a complete work-up including SISCOS and MRI for EZ localization, followed by discussion in the multidisciplinary epilepsy surgery meeting prior to surgery. Ictal & interictal perfusion SPECT studies were performed with Tc-99m Ethylene Cysteinate Dimer (Tc-99m ECD) followed by SISCOS analysis using SPM2 and Bioimage Suite 2.6. Concordance for localization was determined by comparing with the surgical resection site and post-surgical outcomes were assessed using the ILAE classification. The concordance for EZ localization demonstrated by SISCOS was 73.8% and MRI was 82.1%. 52 patients (61.9%) had good surgical outcome and 31(59%) of these were FCD type 2. In patients with discordant MRI findings, SISCOS was able to provide localisation in 86% (13/15), with 69.2% showing good surgical outcomes. Sensitivity of SISCOS and MRI was 73% (95% CI = 59-84.8%) and 78% (95% CI = 67.5-90.3%) respectively with no significant difference between the two. In FCD type I, both SISCOS and MRI revealed a similar a sensitivity of 76.4% (95%CI = 50.1-93.2%). Concordant cases exhibited higher seizure-free odds ratios for both modalities. SISCOS is effective in localizing the EZ in FCD patients, comparable to MRI. Integrating SISCOS and MRI enhances lesion detection, especially in MRI discordant cases. A comprehensive diagnostic approach utilizing SISCOS and MRI can optimize the non-invasive pre-surgical assessment in DRE thereby guiding surgical decision-making in a resource-limited setting.

Endoscopic approaches to the management of dysplasia in inflammatory bowel disease: A state-of-the-art narrative review.

Patients with inflammatory bowel disease (IBD) are at an increased risk of developing colitis-associated neoplasia (CAN), including colorectal cancer (CRC), through the inflammation-dysplasia-neoplasia pathway. Dysplasia is the most reliable, early and actionable marker for CAN in these patients. While such lesions are frequently encountered, adequate management depends on an accurate assessment, complete resection and close surveillance. With recent advances in endoscopic technologies and research in the field of CAN, the management of dysplastic lesions has significantly improved. The American Gastroenterology Association and Surveillance for Colorectal Endoscopic Neoplasia Detection (SCENIC) provide a guideline framework for approaching dysplastic lesions in patients with IBD. However, there are significant gaps in these recommendations and real-world clinical practice. Accurate lesion assessment remains pivotal for adequate management of CAN. Artificial intelligence-guided modalities are now increasingly being used to aid the detection of these lesions further. As the lesion detection technologies are improving, our armamentarium of resection techniques is also expanding and includes hot or cold polypectomy, endoscopic mucosal resection, endoscopic sub-mucosal dissection and full-thickness resection. With the broadened scope of endoscopic resection, the recommendations regarding surveillance after resection has also changed. Certain patient populations such as those with invisible dysplasia or with prior colectomy and ileal pouch anal anastomosis need special consideration. In the present review, we aim to provide a state-of-the-art summary of the current practice of endoscopic detection, resection and surveillance of dysplasia in patients with IBD and provide some perspective on the future directions based on the latest research.

Non-polypoid Colorectal Lesions Detection and False Positive Detection by Artificial Intelligence under Blue Laser Imaging and Linked Color Imaging.

Artificial intelligence (AI) with white light imaging (WLI) is not enough for detecting non-polypoid colorectal polyps and it still has high false positive rate (FPR). We developed AIs using blue laser imaging (BLI) and linked color imaging (LCI) to detect them with specific learning sets (LS). The contents of LS were as follows, LS (WLI): 1991 WLI images of lesion of 2-10 mm, LS (IEE): 5920 WLI, BLI, and LCI images of non-polypoid and small lesions of 2-20 mm. LS (IEE) was extracted from videos and included both in-focus and out-of-focus images. We designed three AIs as follows: AI (WLI) finetuned by LS (WLI), AI (IEE) finetuned by LS (WLI)+LS (IEE), and AI (HQ) finetuned by LS (WLI)+LS (IEE) only with images in focus. Polyp detection using a test set of WLI, BLI, and LCI videos of 100 non-polypoid or non-reddish lesions of 2-20 mm and FPR using movies of 15 total colonoscopy were analyzed, compared to 2 experts and 2 trainees. The sensitivity for LCI in AI (IEE) (83%) was compared to that for WLI in AI (IEE) (76%: p=0.02), WLI in AI (WLI) (57%: p<0.01), BLI in AI (IEE) (78%: p=0.14), and LCI in trainees (74%: p<0.01). The sensitivity for LCI in AI (IEE) (83%) was significantly higher than that in AI (HQ) (78%: p<0.01). The FPR for LCI (6.5%) in AI (IEE) were significantly lower than that in AI (HQ) (17.3%: p<0.01). AI finetuned by appropriate LS detected non-reddish and non-polypoid polyps under LCI.

Low-contrast lesion detection in neck CT: a multireader study comparing deep learning, iterative, and filtered back projection reconstructions using realistic phantoms

BackgroundComputed tomography (CT) reconstruction algorithms can improve image quality, especially deep learning reconstruction (DLR). We compared DLR, iterative reconstruction (IR), and filtered back projection (FBP) for lesion detection in neck CT.MethodsNine patient-mimicking neck phantoms were examined with a 320-slice scanner at six doses: 0.5, 1, 1.6, 2.1, 3.1, and 5.2 mGy. Each of eight phantoms contained one circular lesion (diameter 1 cm; contrast -30 HU to the background) in the parapharyngeal space; one phantom had no lesions. Reconstruction was made using FBP, IR, and DLR. Thirteen readers were tasked with identifying and localizing lesions in 32 images with a lesion and 20 without lesions for each dose and reconstruction algorithm. Receiver operating characteristic (ROC) and localization ROC (LROC) analysis were performed.ResultsDLR improved lesion detection with ROC area under the curve (AUC) 0.724 ± 0.023 (mean ± standard error of the mean) using DLR versus 0.696 ± 0.021 using IR (p = 0.037) and 0.671 ± 0.023 using FBP (p < 0.001). Likewise, DLR improved lesion localization, with LROC AUC 0.407 ± 0.039 versus 0.338 ± 0.041 using IR (p = 0.002) and 0.313 ± 0.044 using FBP (p < 0.001). Dose reduction to 0.5 mGy compromised lesion detection in FBP-reconstructed images compared to doses ≥ 2.1 mGy (p ≤ 0.024), while no effect was observed with DLR or IR (p ≥ 0.058).ConclusionDLR improved the detectability of lesions in neck CT imaging. Dose reduction to 0.5 mGy maintained lesion detectability when denoising reconstruction was used.Relevance statementDeep learning enhances lesion detection in neck CT imaging compared to iterative reconstruction and filtered back projection, offering improved diagnostic performance and potential for x-ray dose reduction.Key PointsLow-contrast lesion detectability was assessed in anatomically realistic neck CT phantoms.Deep learning reconstruction (DLR) outperformed filtered back projection and iterative reconstruction.Dose has little impact on lesion detectability against anatomical background structures.Graphical

FOXP3 serum concentration; a likely predictor of CIN and cervical cancer: Secondary analysis from a case control study at a clinic in South western Uganda

Biomarkers including Forkhead/winged-helix transcription factor box P3 have been proposed in immunohistochemical techniques to diagnose cervical lesions, but can be objectively quantified and measured in blood using methods that can be standardised. In this study we quantified the serum FOXP3 concentrations and assessed their association with cervical lesions at the cervical cancer clinic of Mbarara Regional Hospital (MRRH) Southwestern Uganda.We performed secondary analysis on archived serum samples from a previous unmatched case control study in which we recruited 90 cervical cancer (CC) cases, 90 cervical intraepithelial neoplasia (CIN) cases before any form of treatment and 90 controls. Clinical and demographic data were recorded. We measured FOXP3 concentrations using quantitative ELISA. We performed descriptive statistics and logistic regression in STATA 17 and took P-values of < 0.05 as statistically significant.The mean concentration of FOXP3 was higher in serum samples from CC cases compared with CIN cases and controls, and this difference was statistically significant (P value < 0.001). More than half (52/90,58 %) of serum samples from CC cases had FOXP3 concentrations greater than 0.0545 ng/ml (P value < 0.001). Increase serum FOXP3 expression was not associated with CIN. Increase in serum FOXP3 concentrations were observed to increase the chances of CC by 2 times (OR: 2.094, P value 0.038, 95 % CI: 1.042–––4.209).Serum FOXP3 is likely associated with cervical lesions especially CC in our study population. Serum FOXP3 testing may be useful in resource limited settings to aid detection of such lesions given the challenges associated with cytology and VIA. We recommend diagnostic utility studies for circulating FOXP3 as a biomarker for detection of cervical cancer.

A machine learning approach for skin lesion classification on iOS: implementing and optimizing a convolutional transfer learning model with Create ML

The integration of machine learning (ML) into mobile applications presents unique challenges, particularly in resource-constrained environments such as iOS devices. Skin lesion classification is a critical task in dermatology, where accurate and efficient diagnostic tools can significantly aid in early detection of malignant lesions. This study aims to implement a machine learning-based iOS mobile application and develop a binary classification model for skin lesion images to determine whether a lesion is malignant. The research utilized Create ML to develop a convolutional neural network optimized for iOS, employing a transfer learning approach. A logistic regression model was cascaded with the convolutional neural network N to enhance classification accuracy. The model's performance was assessed through various validation metrics, ensuring its robustness and efficiency within the constraints of mobile hardware. A curated dataset from the International Skin Imaging Collaboration archive was used for training and testing. The model achieved an accuracy of 92%, a precision of 90%, a recall of 93%, and an F1-score of 91% in classifying skin lesions. These metrics validate the model's efficacy in identifying malignant lesions. Data curation involved collecting, labeling, and preparing a dataset from publicly available sources, ensuring the inclusion of diagnostically relevant features. The final model was integrated into an iOS application using Core ML and Vision frameworks. The developed application demonstrates reliable performance in classifying skin lesions with high accuracy on iOS devices. The inclusion of comprehensive performance metrics justifies the efficacy of the proposed approach. Future work will explore enhancements in model architecture and object detection capabilities to further improve diagnostic precision and application usability.

Investigation of distributed learning for automated lesion detection in head MR images.

In this study, we investigated the application of distributed learning, including federated learning and cyclical weight transfer, in the development of computer-aided detection (CADe) software for (1) cerebral aneurysm detection in magnetic resonance (MR) angiography images and (2) brain metastasis detection in brain contrast-enhanced MR images. We used datasets collected from various institutions, scanner vendors, and magnetic field strengths for each target CADe software. We compared the performance of multiple strategies, including a centralized strategy, in which software development is conducted at a development institution after collecting de-identified data from multiple institutions. Our results showed that the performance of CADe software trained through distributed learning was equal to or better than that trained through the centralized strategy. However, the distributed learning strategies that achieved the highest performance depend on the target CADe software. Hence, distributed learning can become one of the strategies for CADe software development using data collected from multiple institutions.

Contemporary algorithms for diagnosing obstructive coronary artery disease in real clinical practice

Background. Despite the high evidence level of the currently existing international recommendations on stable coronary heart disease (CHD) and chronic coronary syndrome, their implementation in domestic clinical practice is insufficient.The aim of the work. To analyze the choice of diagnostic tactics (non-invasive and invasive) in patients with suspected obstructive coronary heart disease in real clinical practice.Methods. The study included outpatients with suspected obstructive CHD, in whom the pre-test probability (PTP) of obstructive CHD was determined; if PTP = 5–15 %, clinical probability was assessed based on CHD risk factors. Based on the results of coronary angiography, the following groups were identified: group I – obstructive lesion of the coronary arteries (≥ 70 %) (n = 50); group II – non-obstructive lesion of the coronary arteries (&lt; 70 %) (n = 32); group III – intact coronary arteries (n = 40). Results. According to the results of coronary angiography, the frequency of detection of obstructive lesion of the coronary arteries was 42 % (in patients without past medical history of myocardial infarction – 31 %). Before performing coronary angiography, non-invasive tests were performed in 2.5 % of cases. Pain in the chest was represented by typical angina in 74 % of patients, with no difference in frequency in all groups. PTP values were statistically significantly higher in the group with obstructive CHD (median – 32 %), however, in the other two groups, PTP values corresponded to a high risk of obstructive CHD (median – 27 % and 21 %, respectively). PTP was an independent predictor for obstructive CHD and subsequent myocardial revascularization.Conclusion. In the cohort of outpatients with suspected coronary heart disease we examined during invasive coronary angiography, the frequency of obstructive lesion of the coronary arteries remains low. Non-invasive tests were performed in isolated cases, while PTP was an independent predictor for obstructive CHD and subsequent myocardial revascularization. To increase the frequency of detection of obstructive coronary heart disease, we should adhere to the diagnostic algorithms of the European Society of Cardiology and make wider use of non-invasive imaging tests.

Somatostatin receptors in fibrotic myocardium.

A patient with a neuroendocrine tumor and history of coronary artery disease underwent PET with 68Ga-DOTATATE PET tracer for tumor visualization. Analysis of the scan showed uptake of 68Ga-DOTATATE in the left ventricle corresponding to previous myocardial infarct. 68Ga-DOTATATE binds by somatostatin receptors (SSTR) and it has been proposed that it may be useful for the detection of cardiac inflammatory lesions. We aimed to test whether SSTR could be upregulated in cardiac fibrotic scar. We analyzed SSTR in cardiac samples from patients with end-stage ischemic cardiomyopathy (ICM, n = 8) and control hearts (n = 5). In mature ICM tissue, SSTR1 and SSTR2 expression was unchanged and SSTR5 expression was significantly decreased in ICM samples vs. control. Immunohistochemistry showed increased SSTR1 and SSTR2 in ICM. Areas with SSTR1 or SSTR2 staining were often adjacent to fibrotic areas. The majority of SSTR1 and SSTR2 staining localized in cardiomyocytes in fibrotic scar-rich areas where CD68 macrophage staining was not present. SSTR are occasionally upregulated in cardiac fibrotic areas. When using 68Ga-DOTATATE PET tracer to detect cardiac sarcoidosis or atherosclerotic plaque, the possibility of tracer uptake in fibrotic areas should be considered.

A systematic review of deep learning-based spinal bone lesion detection in medical images.

Spinal bone lesions encompass a wide array of pathologies, spanning from benign abnormalities to aggressive malignancies, such as diffusely localized metastases. Early detection and accurate differentiation of the underlying diseases is crucial for every patient's clinical treatment and outcome, with radiological imaging being a core element in the diagnostic pathway. Across numerous pathologies and imaging techniques, deep learning (DL) models are progressively considered a valuable resource in the clinical setting. This review describes not only the diagnostic performance of these models and the differing approaches in the field of spinal bone malignancy recognition, but also the lack of standardized methodology and reporting that we believe is currently hampering this newly founded area of research. In line with their established and reliable role in lesion detection, this publication focuses on both computed tomography and magnetic resonance imaging, as well as various derivative modalities (i.e. SPECT). After conducting a systematic literature search and subsequent analysis for applicability and quality using a modified QUADAS-2 scoring system, we confirmed that most of the 14 identified studies were plagued by major limitations, such as insufficient reporting of model statistics and data acquisition, a lacking external validation dataset, and potentially biased annotation. Although we experienced these limitations, we nonetheless conclude that the potential of these methods shines through in the presented results. These findings underline the need for more stringent quality controls in DL studies, as well as model development to afford increased insight and progress in this promising novel field.

Descriptive evaluation of a camera-based dairy cattle lameness detection technology

Lameness in dairy cattle is a clinical sign of impaired locomotion, mainly caused by painful foot lesions, compromising the US dairy industry's economic, environmental, and social sustainability goals. Combining technology and on farm data may be a more precise and less labor-intensive lameness detection tool, particularly for early detection. The objective of this observational study was to describe the association between average weekly autonomous camera-based (AUTO) locomotion scores and hoof trimming (HT) data. The AUTO data were collected from 3 farms from April 2022 to March 2023. Historical farm HT data were collected from March 2016 to March 2023 and used to determine cow lesion history and date of HT event. The HT events were categorized as a regular HT (TRIM; n = 2290) or a HT with a lesion recorded (LESION; n = 670). Events with LESION were sub-categorized based on lesion category: digital dermatitis (DD; n = 276), sole ulcer (SU; n = 79), white line disease (WLD; n = 141), and other (n = 174). The data also contained the leg of the LESION, classified as front left (FL; n = 54), front right (FR; n = 146), rear left (RL; n = 281), or rear right (RR; n = 183) leg with 6 events missing the leg. Cows' HT histories were classified as follows: cows with no previous recorded instance of any lesion were classified as TRIM0 (n = 1554). The first instance of any hoof lesion was classified as LESION1 (n = 238). This classification was retained until a subsequent TRIM occurred - recorded as TRIM1 (n = 632). The next unique instance of any lesion following a TRIM1 was classified as LESION2 (n = 86). Any LESION events occurring after LESION1 or LESION2 without a subsequent TRIM were considered a hoof lesion recurrence and classified as LESIONRE1 (n = 164) and LESIONRE2 (n = 22), respectively. TRIM events after LESION2 or LESION2RE (n = 104) or LESION events after LESIONRE1 or LESIONRE2 were classified as LESION_OTHER (n = 160). The AUTO scores from -28 to -1 days prior to the HT event were summarized into weekly scores and included if cows had at least 1 observation per week in the 4 weeks before the event. For all weeks, LESION cows had a higher median AUTO score than TRIM cows. Cows with TRIM0 had the lowest and most consistent median weekly score compared to LESION and other TRIM classifications. Before HT cows with TRIM0 and TRIM1, both had median score increases of 1 across the 4 weeks, while the LESION categories had an increase of 4 to 8. Scores increased with each subsequent LESION event compared to the previous LESION event. Cows with SU lesions had the highest median score across the 4 weeks, WLD had the largest score increase, and DD had the lowest median score and score increase. When grouping a LESION event by leg the hoof lesion was found on, the AUTO scores for four groups displayed comparable median values. Due to the difference between TRIM and LESION events, this technology shows potential for the early detection of hoof lesions.

Exploring deep learning strategies for intervertebral disc herniation detection on veterinary MRI

Intervertebral Disc Herniation (IVDH) is a common spinal disease in dogs, significantly impacting their health, mobility, and overall well-being. This study initiates an effort to automate the detection and localization of IVDH lesions in veterinary MRI scans, utilizing advanced artificial intelligence (AI) methods. A comprehensive canine IVDH dataset, comprising T2-weighted sagittal MRI images from 213 pet dogs of various breeds, ages, and sizes, was compiled and utilized to train and test the IVDH detection models. The experimental results showed that traditional two-stage detection models reliably outperformed one-stage models, including the recent You Only Look Once X (YOLOX) detector. In terms of methodology, this study introduced a novel spinal localization module, successfully integrated into different object detection models to enhance IVDH detection, achieving an average precision (AP) of up to 75.32%. Additionally, transfer learning was explored to adapt the IVDH detection model for a smaller feline dataset. Overall, this study provides insights into advancing AI for veterinary care, identifying challenges and exploring potential strategies for future development in veterinary radiology.

Optimizing computed tomography image reconstruction for focal hepatic lesions: Deep learning image reconstruction vs iterative reconstruction

BackgroundDeep learning image reconstruction (DLIR) is a novel computed tomography (CT) reconstruction technique that minimizes image noise, enhances image quality, and enables radiation dose reduction. This study aims to compare the diagnostic performance of DLIR and iterative reconstruction (IR) in the evaluation of focal hepatic lesions. MethodsWe conducted a retrospective study of 216 focal hepatic lesions in 109 adult participants who underwent abdominal CT scanning at our institution. We used DLIR (low, medium, and high strength) and IR (0 %, 10 %, 20 %, and 30 %) techniques for image reconstruction. Four experienced abdominal radiologists independently evaluated focal hepatic lesions based on five qualitative aspects (lesion detectability, lesion border, diagnostic confidence level, image artifact, and overall image quality). Quantitatively, we measured and compared the level of image noise for each technique at the liver and aorta. ResultsThere were significant differences (p < 0.001) among the seven reconstruction techniques in terms of lesion borders, image artifacts, and overall image quality. Low-strength DLIR (DLIR-L) exhibited the best overall image quality. Although high-strength DLIR (DLIR-H) had the least image noise and fewest artifacts, it also had the lowest scores for lesion borders and overall image quality. Image noise showed a weak to moderate positive correlation with participants’ body mass index and waist circumference. ConclusionsThe optimal-strength DLIR significantly improved overall image quality for evaluating focal hepatic lesions compared to the IR technique. DLIR-L achieved the best overall image quality while maintaining acceptable levels of image noise and quality of lesion borders.

Diagnostic ability of [18F]FDG PET/CT for distinguishing benign from malignant spleen lesions.

[18F]Fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) is a non-invasive imaging modality used in the differential diagnosis of splenic lesions, although ideal parameters and thresholds remain unclear. The present study evaluated the ability of [18F]FDG PET/CT, including its visual and quantitative parameters, to differentiate between benign and malignant splenic lesions. Patients who underwent [18F]FDG PET/CT following the detection of splenic lesions on contrast-enhanced CT were retrospectively analysed. Visual parameters assessed on [18F]FDG PET/CT included whole spleen uptake intensity, lesion multiplicity, and lesion uptake, and quantitative parameters included maximum standardised uptake value (SUVmax), lesion-to-background ratio (LBR), metabolic tumour volume (MTV), total lesion glycolysis (TLG), and lesion size. Parameters differentiating between benign and malignant lesions were evaluated by Pearson's chi-square test, Mann-Whitney U-test, and receiver operating characteristics (ROC) curve analysis. Splenic lesion uptake (p = 0.001) was the only visual parameter significantly distinguishing between benign and malignant lesions. ROC curve analysis demonstrated that SUVmax had the largest area under the ROC, 0.91 (p < 0.001), with an optimal cut-off > 5.3 having a sensitivity of 90.3% and a specificity of 80.6%. Subgroup analysis of malignant lesions showed that SUVmax (p = 0.013), LBR (p = 0.012), and TLG (p = 0.034) were significantly higher in splenic lymphomas than in splenic metastases. Of the [18F]FDG PET/CT parameters investigated, SUVmax had the highest accuracy in diagnosing malignant splenic lesions and was significantly higher in splenic lymphomas than in splenic metastases. Visual determination of [18F]FDG uptake by splenic lesions may be an easily evaluated parameter. SUVmax and visual grade of [18F]FDG PET/CT help to differentiate spleen lesions. [18F]FDG PET/CT is useful for discriminating between benign and malignant spleen lesions. Many splenic lesions are difficult to diagnose on anatomical imaging, with histopathologic analyses are required. SUVmax of PET/CT provided the diagnostic ability to differentiate between benign and malignant splenic lesions. More than normal spleen uptake can be a convenient parameter to diagnose malignant spleen lesions.

Detection of myeloma-associated osteolytic bone lesions with energy-integrating and photon-counting detector CT.

Arecent innovation in computed tomography (CT) imaging has been the introduction of photon-counting detector CT (PCD-CT) systems, which are able to register the number and the energy level of incoming x‑ray photons and have smaller detector elements compared with conventional CT scanners that operate with energy-integrating detectors (EID-CT). The study aimed to evaluate the potential benefits of anovel, non-CE certified PCD-CT in detecting myeloma-associated osteolytic bone lesions (OL) compared with astate-of-the-art EID-CT. Nine patients with multiple myeloma stageIII (according to Durie and Salmon) underwent magnetic resonance imaging (MRI), EID-CT, and PCD-CT of the lower lumbar spine and pelvis. The PCD-CT and EID-CT images of all myeloma lesions that were visible in clinical MRI scans were reviewed by three radiologists for corresponding OL. Additionally, the visualization of destructions to cancellous or cortical bone, and trabecular structures, was compared between PCD-CT and EID-CT. Readers detected 21% more OL in PCD-CT than in EID-CT images (138 vs. 109; p < 0.0001). The sensitivity advantage of PCD-CT in lesion detection increased with decreasing lesion size. The visualization quality of cancellous and cortical destructions as well as of trabecular structures was rated higher by all three readers in PCD-CT images (mean image quality improvements for PCD-CT over EID-CT were +0.45 for cancellous and +0.13 for cortical destructions). For myeloma-associated OL, PCD-CT demonstrated significantly higher sensitivity, especially with small size. Visualization of bone tissue and lesions was considered significantly better in PCD-CT than in EID-CT. This implies that PCD-CT scanners could potentially be used in the early detection of myeloma-associated bone lesions.