Diagnosis Of Focal Liver Lesions Research Articles

Robust End-to-End Focal Liver Lesion Detection Using Unregistered Multiphase Computed Tomography Images

The computer-aided diagnosis of focal liver lesions (FLLs) can help improve workflow and enable correct diagnoses; FLL detection is the first step in such a computer-aided diagnosis. Despite the recent success of deep-learning-based approaches in detecting FLLs, current methods are not sufficiently robust for assessing misaligned multiphase data. By introducing an attention-guided multiphase alignment in feature space, this study presents a fully automated, end-to-end learning framework for detecting FLLs from multiphase computed tomography (CT) images. Our method is robust to misaligned multiphase images owing to its complete learning-based approach, which reduces the sensitivity of the model’s performance to the quality of registration and enables a standalone deployment of the model in clinical practice. Evaluation on a large-scale dataset with 280 patients confirmed that our method outperformed previous state-of-the-art methods and significantly reduced the performance degradation for detecting FLLs using misaligned multiphase CT images. The robustness of the proposed method can enhance the clinical adoption of the deep-learning-based computer-aided detection system.

Applications of Artificial Intelligence in the Automatic Diagnosis of Focal Liver Lesions: A Systematic Review

Focal liver lesions (FLLs) are defined as abnormal solid or liquid masses differentiated from normal liver, frequently being clinically asymptomatic. The aim of this systematic review is to provide a comprehensive overview of current artificial intelligence (AI) applications, deep learning systems and convolutional neural networks, capable of performing a completely automated diagnosis of FLLs. We searched PubMed, Cochrane Library, EMBASE, and WILEY databases using predefined keywords. Articles were screened for relevant publications about AI applications capable of automated diagnosis of FLLs. The search terms included: (focal liver lesions OR FLLs OR hepatic focal lesions OR liver focal lesions OR liver tumor OR hepatic tumor) AND (artificial intelligence OR machine learning OR neural networks OR deep learning OR automated diagnosis OR ultrasound OR US OR computer scan OR CT OR magnetic resonance imaging OR MRI OR computer-aided diagnosis OR automated computer tomography OR automated magnetic imaging). Our search identified a total of 32 articles analyzing complete automated imagistic diagnosis of FLLs, out of which 14 studies analyzing liver ultrasound images, 8 studies analyzing computer tomography images and 10 studies analyzing images obtained from magnetic resonance imaging. We found significant evidence demonstrating that implementing a complete automated system for FLLs diagnosis using AI-based applications is currently feasible. Various automated AI-based applications have been analyzed. However, there is no clear evidence about the superiority of any of the systems.

Application of diffusion kurtosis imaging in differential diagnosis of focal liver lesions.

Diffusion kurtosis imaging (DKI) is an MRI method related to diffusion imaging (DWI) that is distinguished by a non-Gaussian calculation of water particles movements in tissues. The aim of the study was to assess DKI advantage over DWI in differentiating benign and malignant liver lesions. Analysis included prospectively acquired group of 83 patients referred consecutively for 3T-MRI liver tumor examination, with 95 liver lesions (31 benign, 59 malignant). MRI assessments were performed with standard protocol and DKI sequence with seven b-values (0-2,000 s/mm2). Quantitative data were acquired by placing ROIs in liver tumors on all b-value images, ROI data extracted, and calculation of DWI and DKI parameters. ADC was calculated for all b-values (ADC0-2000) and for three values of b = 0, 500, and 750 (s/mm2) (ADC0-500-750). DKI and ADC parameters for benign and malignant lesions were compared, and ROC curves were plotted. Significant differences were obtained for all DKI and ADC parameters. ROC analysis showed AUC of DK, K, ADC0-2000, and ADC0-500-750 was 0.74, 0.77, 0.77, and 0.75, respectively. The highest sensitivity (of 0.91) was obtained for ADC0-2000. The highest specificity (0.65) and accuracy (0.80) was obtained for K. DKI technique yields statistically comparable results with DWI technique.

Strategy for Accurate Diagnosis by Contrast-Enhanced Ultrasound of Focal Liver Lesions in Patients Not at High Risk for Hepatocellular Carcinoma: A Preliminary Study.

To develop an effective strategy for accurate diagnosis of focal liver lesions (FLLs) in patients with non-high risk for hepatocellular carcinoma (HCC). From January 2012 to December 2015, consecutive patients with non-high risk for HCC who underwent contrast-enhanced ultrasound (CEUS) were included in this retrospective double-reader study. All patients were stratified into 2 different risks (intermediate, low-risk) groups according to criteria based on clinical characteristics, known as clinical risk stratification criteria. For the intermediate-risk group, the CEUS criteria for identifying benign lesions and HCCs were constructed based on selected CEUS features. The diagnostic performance of the clinical risk stratification criteria, and CEUS criteria for identifying benign lesions and HCCs was evaluated. This study included 348 FLLs in 348 patients. The sensitivity and specificity of the clinical risk stratification criteria for malignancy was 97.8 and 69.8%. Patients were classified as intermediate risk if they were male, or older than 40 years of age, or HBcAb positive, or having positive tumor markers. Otherwise, patients were classified as low risk. Among the 348 patients, 327 were in the intermediate-risk group and 21 were in the low-risk group. In the intermediate-risk group, the CEUS criteria for identifying benign lesions were any of the following features: 1) hyper/isoenhancement in the arterial phase without washout, 2) nonenhancement in all phases, 3) peripheral discontinuous globular enhancement in the arterial phase, 4) centrifugal enhancement or peripheral enhancement followed by no central enhancement, or 5) enhanced septa. The accuracy, sensitivity, and specificity of the CEUS criteria for identifying benign lesions were 94.5, 83.0, and 99.6%, respectively. Arterial phase hyperenhancement followed by mild and late washout (>60 seconds) was more common in HCC patients than in non-HCC patients (P < .001). Using arterial phase hyperenhancement followed by mild and late washout as the CEUS criteria for identifying HCCs, the sensitivity and specificity were 52.6 and 95.3%, but unfortunately, the positive predictive value was only 82.0%. For the low-risk group, no further analysis was performed due to the small sample size. Initial clinical risk stratification followed by assessment of certain CEUS features appears to be a promising strategy for the accurate diagnosis of FLLs in patients not at high risk for HCC.

Diagnostic Accuracy of Triphasic CT in the Characterization of Incidental Focal Liver Lesions

Objectives: To determine the diagnostic accuracy of triphasic CT in the diagnosis of focal liver lesions as benign or malignant by using histopathology as the gold standard. Methods: This descriptive cross-sectional study was carried out at the diagnostic radiology department of the Pakistan Institute of Medical Sciences (P.I.M.S.) in Islamabad.110 individuals with localized liver lesions between the ages of 15 and 65 years and both genders were included. All of the study subjects were then undergoing triphasic CT scan of the liver. The CT findings were recorded as benign or malignant and correlated with histopathology findings. Data collection was done by using the self-structured proforma. Results: Patient’s mean age was 43.92 ± 12.87 years. Out of all, males were 68 (61.82%) and females were 42 (38.18%), with a female-to-male ratio of 1:1.6. Malignant lesions were 73 (66.36%) and 37 (33.54%) patients had benign lesions as per Triphasic CT diagnosis. Furthermore, in these 73 malignant cases, 67 cases were observed with malignant lesions and 6 cases had benign lesions on histopathology, while among the 37 CT cases of benign lesions, 8 cases had malignant lesions as per the histopathological diagnosis. Although triphasic CT showed 89.33% sensitivity and 82.86% specificity in the diagnosis of focal liver lesions by using histopathology as the gold standard. Conclusion: Triphasic CT was observed to be a non-invasive, accurate, and highly sensitive imaging technique in the diagnosis of focal liver lesions, whether (benign or malignant). Therefore, it can be used frequently to lower the number of pure diagnostic biopsies. Keywords: Malignant hepatic lesions, non-invasive, imaging modality, sensitivity

Endoscopic Ultrasound-Guided Fine-Needle Biopsy versus Fine-Needle Aspiration in the Diagnosis of Focal Liver Lesions: Prospective Head-to-Head Comparison.

Endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) or fine-needle aspiration (EUS-FNA) from focal liver lesions are indicated in selected cases, but there has been no previous comparison of needle types of the same size. The aim of our study was to compare the histologic diagnostic accuracy and adequacy of cores obtained with EUS-FNB needles in contrast to those obtained with FNA needles in focal liver lesions. This prospective one-center study included patients with left lobe hepatic focal lesions with contraindications for percutaneous liver biopsy or need for EUS for concomitant lesions. Each patient had one pass of 22G EUS-FNB (Franseen) needle and one pass of 22G EUS-FNA in a crossover manner, without macroscopic on-site evaluation. Each sample was analyzed separately for histologic adequacy and diagnosis. The final diagnosis was based on histology results or on imaging follow-up in the case of negative biopsies. The EUS-FNB samples (n = 30) were found to be more adequate for histologic analysis, with more cellularity and longer tissue aggregates than the EUS-FNA samples (n = 30). The accuracy of EUS-FNB was 100%, whereas that of EUS-FNA was 86.7% (p = 0.039). No post-procedure complications were noted. The 22G EUS-FNB needle proved superior to 22G EUS-FNA in terms of tissue acquisition diagnostic accuracy and histologic adequacy in focal liver lesions.

Application of Diffusion-Weighted Imaging and Hepatobiliary-Specific Contrast Agent Gd-EOB-DTPA in the Diagnosis and Differential Diagnosis of Focal Liver Lesions

There are many types of focal liver lesions (FLL) presenting different lesion signs and their diagnosis and differential diagnosis are relatively difficult. It is of great clinical significance to accurately detect, classify and characterize focal liver lesions as soon as possible. Diffusion-weighted imaging (DWI) provides information on liver cell density, microstructure, and microcirculation perfusion. Gadolinium-ethoxibenzyl-diethylenetriamine pentaacetic acid (Gd-EOB-DTPA) is a hepatobiliary-specific contrast agent. Gd-EOB-DTPA-enhanced MRI examination of liver provides information on the blood perfusion of lesions and specific information on the uptake function of normal liver cells. The combined application of the two can significantly improve the sensitivity and diagnostic accuracy in the detection of FLL. Herein, we reviewed the research findings on the application of DWI and Gd-EOB-DTPA in FLL diagnosis in order to provide reference for further clinical application. Most of the existing studies only made comparison and discussion of the DWI image quality of different b values and their fitted apparent diffusion coefficient (ADC) values before and after Gd-EOB-DTPA enhancement, and the reported findings are not only varied, but also inconsistent. Whether Gd-EOB-DTPA will affect DWI images is still been debated. Future research should focus on quantitative comparison, discussion and verification of the enhancement effect after injection of Gd-EOB-DTPA, as well as the changes in the ADC value corresponding to different b values before and after enhancement, in order to provide more objective and consistent research results for clinical application.

Analysis of the Role of Shear Wave Elastography in Diagnosing Focal Liver Lesions

Objective: Focal hepatic lesions are the second leading cause of death in men worldwide. However, shear wave elastography (SWE) has proved to help assess liver fibrosis. We aim to demonstrate the role of shear wave elastography in the diagnosis of focal liver lesions. Methods: A prospective study was conducted in the radiology department of CPEIC Multan from April 2021-April 2022. A total of 70 patients with 93 focal lesions were included. Shear wave elastography (SWE) was performed on all the patients, and local lesions and parenchyma stiffness values were calculated. Ten patients were excluded as they failed elastography acquisitions. Contrast-enhanced CT and MRI were performed on the remaining 60 patients. Thirty-two patients underwent liver biopsy. Benign focal lesions were confirmed by analyzing the results of ultrasound biopsy, CT, and MRI. Results: Cholangiocarcinoma was found to have the highest stiffness value (34.2kPa), hence the stiffest malignant lesion. Focal nodular hyperplasia had a stiffness value of 25.4kPa and was the most stiff benign lesion. The average stiffness value between malignant and benign lesions had a significant difference, the value of malignant lesions being significantly high ( p&lt;0.001). Conclusion: Shear wave elastography can efficiently differentiate between malignant and benign hepatic lesions and can individually characterize these lesions accurately. Keywords: Focal liver lesions, malignant, stiffness value, shear wave elastography, benign

HCC or Something Else? Frequency of Various Benign and Malignant Etiologies in Cirrhotic Patients with Newly Detected Focal Liver Lesions in Relation to Different Clinical and Sonographic Parameters.

Background and Aims: To investigate the frequency of different benign and malignant focal liver lesions (FLLs) in relation to clinical and sonographic features among patients with liver cirrhosis (LC) and newly detected FLLs. Methods: This study was a retrospective analysis of 225 cirrhotic patients with newly detected FLLs who underwent hepatic ultrasound (US) examinations at our university hospital from 2011 to 2022. The diagnosis of FLLs was based on histology and/or consensus radiological criteria, in accordance with the current diagnostic guidelines. The FLLs were classified into benign (bFLLs) or malignant (mFLLs) lesions and the latter group was subclassified into HCC and non-HCC mFLLs. The frequency, clinical parameters, and sonographic features of the different groups were examined and compared. Results: Of the 225 FLLs, 154 (68.4%) were mFLLs and 71 (31.6%) bFLLs. HCC was the most frequent subcategory of FLLs (132; 58.7%). There were (22; 9.8%) non-HCC mFLLs with 11 (4.9%) metastases and 11 (4.9%) non-HCC primary liver tumors. Regenerative nodules (RNs) were the most frequent form of bFLLs (25; 11.1%), followed by simple cysts (22; 9.8%) and hemangiomas (14; 6.2%). The other bFLLs (10; 14.1%) were fat deposition/sparing (5), hematomas (2), abscesses (2), and echinococcal cysts (1). The distribution of bFLLs and HCC and non-HCC mFLLs varied significantly according to the clinical scenarios. HCC mFLLs were more frequent in males (p = 0.001), in those with no history of active non-hepatic primary malignant disease (NHPMD) (p < 0.001), in those with a hepatitis B or C etiology of LC (p = 0.002), when located in the right lobe (p = 0.008), and when portal vein thrombosis was present (p = 0.03). Conclusion: In cirrhotic patients with newly detected FLLs, the non-HCC etiology was more frequently diagnosed in lesions that were located in the left lobe, in females, and in patients with a history of active NHPMD. Thus, the lower frequency of HCC in the abovementioned groups demonstrated that a cautious implementation of the current consensus radiological criteria would be required for these groups, particularly in patients with an active NHPMD, given the fact that the consensus criteria were not validated in these populations. A more active diagnostic approach may ultimately be needed for these patients. Large prospective studies are needed to validate these findings.

3D variable flip angle T1 mapping for differentiating benign and malignant liver lesions at 3T: comparison with diffusion weighted imaging

BackgroundDifferent methods have been used to improve the imaging diagnosis of focal liver lesions (FLL). Among them, magnetic resonance imaging (MRI) has received more attention since it provides significant amount of information without radiation exposure. However, atypical imaging characteristics of FLL on MRI may complicate the differential diagnosis between benign and malignant FLL. This study aimed to compare the diagnostic value of T1 mapping and diffusion-weighted imaging (DWI) for differentiating of benign and malignant FLLs.MethodsThis retrospective study enrolled 294 FLLs, including 150 benign and 144 malignant lesions. Whole liver T1 mapping sequences were obtained before and 2 min after the administration of Gd-DTPA to acquire native T1 and enhanced T1 and ΔT1%. Additionally, DWI sequence was conducted to generate apparent diffusion coefficient (ADC) maps. These quantitative parameters were compared using one-way analysis of variance, and the diagnostic accuracy of T1 mapping and ADC for FLLs was calculated by area under the curve (AUC).ResultsSignificant differences were observed regarding the native T1, enhanced T1, ΔT1%, and ADC between benign and malignant FLLs. Furthermore, the sensitivity and specificity of the parameters are as follows: native T1 0.797/0.702 (cut off value 1635.5 ms); enhanced T1, 0.911/0.976 (cutoff value 339.2 ms); ΔT1%, 0.901/0.905 (cutoff value 70.8%); and ADC, 0.975/0.952 (cutoff value 1.21 × 10−3 mm2/s). The ideal cutoff values for native T1 and ADC in identifying cyst and haemangioma were 2422.9 ms (AUC 0.990, P < 0.01) and 2.077 × 10–3 mm2/s (AUC 0.949, P < 0.01), respectively, with a sensitivity and specificity of 0.963/1 and 0.852/0.892, respectively. ADC was significantly positively correlated with T1 and ΔT1%, and significantly negatively correlated with enhanced T1.ConclusionThe 3D Variable flip angle T1 mapping technique with Gd-DTPA has a high clinical potential for identifying benign and malignant FLLs. The enhanced T1 and ΔT1% values have similar diagnostic accuracy compared with DWI in evaluating FLLs. Native T1 shows better performance than DWI in distinguishing benign liver lesions, specifically, cysts, and haemangioma.

Contrast-Enhanced Ultrasound and Magnetic Resonance Enhancement Based on Machine Learning in Cancer Diagnosis in the Context of the Internet of Things Medical System.

As an accurate, safe, and effective noninvasive examination method, imaging examination has been widely used in the diagnosis and differential diagnosis of focal liver lesions. Enhanced ultrasonography (CEUS), enhanced CT (CECT), and enhanced magnetic resonance imaging (CEMRI) are the most commonly used enhanced imaging methods in clinical practice, all of which can accurately determine the nature of liver lesions. The purpose of this paper is to study the application of contrast-enhanced ultrasound and magnetic resonance enhancement in cancer diagnosis based on the Internet of Things medical system. The basic clinical data, CEUS, and enhanced CT/MRI findings of 120 CHC patients were retrospectively analyzed. The clinicopathological features of CHC patients were investigated by contrast-enhanced ultrasonography and CT/MRI enhanced mode. The diagnostic value of contrast-enhanced ultrasound and enhanced CT/MRI combined with tumor markers in CHC was analyzed. The experimental results showed that the sensitivities of CEUS, enhanced MRI, and their combination in diagnosing CHC were 72.44%, 81.56%, and 93.78%, respectively. This experiment has an important value in the diagnosis of primary liver cancer.

COMPARATIVE EVALUATION OF ULTRASONOGRAPHY AND COMPUTED TOMOGRAPHY IN FOCAL LIVER LESIONS

Introduction: Liver diseases are among the commonestcauses of morbidity and mortality and accurate characterization is an important goal of diagnostic imaging.The aim of the study is to establish the role of ultrasonography and CT in differentiation and early diagnosis of focal liver lesions and to correlate the findings with pathological diagnosis. Methodology:This prospective comparative observational and hospital based study was conducted on 100 patients. USG and CT were performed on focal hepatic mass patients. Diagnostic value of USG were compared with to those of CT. Results: Ultrasonography and CT evaluation results were compared with final diagnosis by FNAC/ HPR.USG showed maximum cases among malignant lesions were HCC (20%) and among benign lesions were liver abscess (19%).CT diagnosed maximum number of cases as metastasis (19%) in malignant lesions andalmost equal percentage of hemangiomas(15%) and liver abscess(16%) among the benign lesions. The sensitivity and specificity of USG in diagnosing lesions turned out to be slightly less than that of CT. Conclusion:The study concluded that ultrasonography can be adopted as the initial imaging modality with advantages of differentiating cystic from solid lesions, relationship with critical structures and can guide FNAC and biopsy. CT has a comparatively higher sensitivity and specificity in detection of lesions and can help narrow the differentials.

Deep learning radiomics for focal liver lesions diagnosis on long-range contrast-enhanced ultrasound and clinical factors.

Routine clinical factors play an important role in the clinical diagnosis of focal liver lesions (FLLs); however, they are rarely used in computer-assisted diagnosis. Therefore, we developed a deep learning (DL) radiomics model, and investigated its effectiveness in diagnosing FLLs using long-range contrast-enhanced ultrasound (CEUS) cines and clinical factors. Herein, 303 patients with pathologically confirmed FLLs after surgery at three hospitals were retrospectively enrolled and divided into a training cohort (n=203), internal validation (IV) cohort (n=50) from one hospital with the ratio of 4:1, and external validation (EV) cohort (n=50) from the other two hospitals. Four DL radiomics models, namely Four Stream 3D convolutional neural network (FS3DU) (trained with CEUS cines only), FS3DU+A (trained with CEUS cines and alpha fetoprotein), FS3DU+H (trained with CEUS cines and hepatitis), and FS3DU+A+H (trained with CEUS cines, alpha fetoprotein, and hepatitis), were formed based on 3D convolutional neural networks (CNNs). They used approximately 20-s preoperative CEUS cines and/or clinical factors to extract spatiotemporal features for the classification of FLLs and the location of the region of interest. The area under curve of the receiver operating characteristic and diagnosis speed were calculated to evaluate the models in the IV and EV cohorts, and they were compared with those of two radiologists. Two-sided Delong tests were used to calculate the statistical differences between the models and radiologists. FS3DU+A+H, which incorporated CEUS cines, hepatitis, and alpha fetoprotein, achieved the highest area under curve of 0.969 (95% CI: 0.901-1.000) and 0.957 (95% CI: 0.894-1.000) among radiologists and other models in IV and EV cohorts, respectively. A significant difference was observed when comparing FS3DU and radiologist 2 (all P<0.05). The diagnosis speed of all the models was the same (10.76 s per patient), and it was two times faster than those of the radiologists (radiologist 1: 23.74 and 27.75 s; radiologist 2: 25.95 and 29.50 s in IV and EV cohorts, respectively). The proposed DL radiomics demonstrated excellent performance on the benign and malignant diagnosis of FLLs by combining CEUS cines and clinical factors. It could help the individualized characterization of FLLs, and enhance the accuracy of diagnosis in the future.

Effectiveness of ADC histogram analysis in the diagnosis of focal liver lesions; is a contrast agent necessary?

Objective: The diagnostic success of apparent diffusion coefficient (ADC) histogram analysis in focal liver lesions, and the effects of quantitative data added to contrast-enhanced abdominal magnetic resonance imaging (MRI) on the diagnostic accuracy were investigated. Materials and Methods: The conventional MRI findings of 524 lesions in total were retrospectively examined. Contrast enhancement kinetics and ADC values for each lesion were found through an image analysis software. Results: Three hundred and fifty (67%) of the lesions were diagnosed as benign and 174 (33%) as malignant. Statistically significant difference was found between the benign and malignant lesion groups in terms of the minimum, maximum and mean ADC values (p&amp;lt;0.001). When optimal thresholds for minimum, maximum and mean ADC were taken as 1.47x10-3mm2/s; 1.85x10-3mm2/s; 1.72x10-3mm2/s respectively, sensitivity was found to be 97%; 83%; 95%, specificity was 100%; 98%; 99%, NPV was 100%; 99%; 99%, and PPV was 93%; 74%; 90%. ADC values added to MRI increased the diagnostic success for metastases (92%→96%), HCC (63%→73%), hemangioma (90%→99%) and FNH (56%→75%). Conclusion: ADC measurement could not show reasonable success in the diagnosis of specific lesions while being successful in the differentiation of benign and malignant lesions. Minimum ADC is more successful than mean and maximum ADC. A non-contrastenhanced MRI protocol based on the ADC measurement applicable to the selected patient group may be helpful.

VueBox® for quantitative analysis of contrast-enhanced ultrasound in liver tumors1.

Dynamic contrast-enhanced ultrasound (DCE-US) enables quantification of tumor perfusion. VueBox is a platform independent external software using DICOM cine loops which objectively provides various DCE-US parameters of tumor vascularity. This review summaries its use for diagnosis and treatment monitoring of liver tumors. The existing literature provides evidence on the successful application of Vuebox based DCE-US for characterization and differential diagnosis of focal liver lesions, as well as on its use for monitoring of local ablative therapies and of modern systemic treatment in oncology.

Contrast-Enhanced Ultrasound as a Main Radiological Diagnostic Method for Primary Liver Neoplasms and Hemangiomas.

Contrast-enhanced ultrasound (CEUS) is a relatively new approach for the definitive diagnosis of focal liver lesions (FLL). The essential advantages of CEUS are affordability, absence of radiation, and negligible nephrotoxicity-making this diagnostic approach more preferable.This review includes data from 39 different research studies published during the last 10 years, selected through the MeSH strategy in PubMed.We conclude that CEUS is a promising approach for diagnosing primary liver neoplasms and it is an excellent radiological approach for children and pregnant women because of the absence of radiation and nephrotoxicity. Studies showed that CEUS is a very good approach for the differentiation of a variety of hemangiomas and for a detailed description of those findings. Therefore, CEUS is an important and progressive method for the diagnosis of liver neoplasms. The regular use of CEUS will facilitate the diagnosis of primary liver lesions.

Diagnosis of focal liver lesions with deep learning-based multi-channel analysis of hepatocyte-specific contrast-enhanced magnetic resonance imaging.

BACKGROUNDThe nature of input data is an essential factor when training neural networks. Research concerning magnetic resonance imaging (MRI)-based diagnosis of liver tumors using deep learning has been rapidly advancing. Still, evidence to support the utilization of multi-dimensional and multi-parametric image data is lacking. Due to higher information content, three-dimensional input should presumably result in higher classification precision. Also, the differentiation between focal liver lesions (FLLs) can only be plausible with simultaneous analysis of multi-sequence MRI images. AIMTo compare diagnostic efficiency of two-dimensional (2D) and three-dimensional (3D)-densely connected convolutional neural networks (DenseNet) for FLLs on multi-sequence MRI.METHODSWe retrospectively collected T2-weighted, gadoxetate disodium-enhanced arterial phase, portal venous phase, and hepatobiliary phase MRI scans from patients with focal nodular hyperplasia (FNH), hepatocellular carcinomas (HCC) or liver metastases (MET). Our search identified 71 FNH, 69 HCC and 76 MET. After volume registration, the same three most representative axial slices from all sequences were combined into four-channel images to train the 2D-DenseNet264 network. Identical bounding boxes were selected on all scans and stacked into 4D volumes to train the 3D-DenseNet264 model. The test set consisted of 10-10-10 tumors. The performance of the models was compared using area under the receiver operating characteristic curve (AUROC), specificity, sensitivity, positive predictive values (PPV), negative predictive values (NPV), and f1 scores.RESULTSThe average AUC value of the 2D model (0.98) was slightly higher than that of the 3D model (0.94). Mean PPV, sensitivity, NPV, specificity and f1 scores (0.94, 0.93, 0.97, 0.97, and 0.93) of the 2D model were also superior to metrics of the 3D model (0.84, 0.83, 0.92, 0.92, and 0.83). The classification metrics of FNH were 0.91, 1.00, 1.00, 0.95, and 0.95 using the 2D and 0.90, 0.90, 0.95, 0.95, and 0.90 using the 3D models. The 2D and 3D networks' performance in the diagnosis of HCC were 1.00, 0.80, 0.91, 1.00, and 0.89 and 0.88, 0.70, 0.86, 0.95, and 0.78, respectively; while the evaluation of MET lesions resulted in 0.91, 1.00, 1.00, 0.95, and 0.95 and 0.75, 0.90, 0.94, 0.85, and 0.82 using the 2D and 3D networks, respectively.CONCLUSIONBoth 2D and 3D-DenseNets can differentiate FNH, HCC and MET with good accuracy when trained on hepatocyte-specific contrast-enhanced multi-sequence MRI volumes.

Diagnostyka zmian ogniskowych wątroby z zastosowaniem ultrasonograficznych środków kontrastujących – strategia badania

The aim of the paper is to present the technique and strategy of ultrasound examination with the use of ultrasound contrast agents in the diagnosis of focal liver lesions depending on the indications and clinical situation.

Development and validation of artificial intelligence to detect and diagnose liver lesions from ultrasound images.

Artificial intelligence (AI) using a convolutional neural network (CNN) has demonstrated promising performance in radiological analysis. We aimed to develop and validate a CNN for the detection and diagnosis of focal liver lesions (FLLs) from ultrasonography (USG) still images. The CNN was developed with a supervised training method using 40,397 retrospectively collected images from 3,487 patients, including 20,432 FLLs (hepatocellular carcinomas (HCCs), cysts, hemangiomas, focal fatty sparing, and focal fatty infiltration). AI performance was evaluated using an internal test set of 6,191 images with 845 FLLs, then externally validated using 18,922 images with 1,195 FLLs from two additional hospitals. The internal evaluation yielded an overall detection rate, diagnostic sensitivity and specificity of 87.0% (95%CI: 84.3-89.6), 83.9% (95%CI: 80.3-87.4), and 97.1% (95%CI: 96.5-97.7), respectively. The CNN also performed consistently well on external validation cohorts, with a detection rate, diagnostic sensitivity and specificity of 75.0% (95%CI: 71.7-78.3), 84.9% (95%CI: 81.6-88.2), and 97.1% (95%CI: 96.5-97.6), respectively. For diagnosis of HCC, the CNN yielded sensitivity, specificity, and negative predictive value (NPV) of 73.6% (95%CI: 64.3-82.8), 97.8% (95%CI: 96.7-98.9), and 96.5% (95%CI: 95.0-97.9) on the internal test set; and 81.5% (95%CI: 74.2-88.8), 94.4% (95%CI: 92.8-96.0), and 97.4% (95%CI: 96.2-98.5) on the external validation set, respectively. CNN detected and diagnosed common FLLs in USG images with excellent specificity and NPV for HCC. Further development of an AI system for real-time detection and characterization of FLLs in USG is warranted.

Value of Contrast-Enhanced Ultrasound in Differential Diagnosis of Focal Liver Lesions

The article discusses the modern possibilities of identifying inflammatory bowel diseases using methods of radiation diagnosis. Methods of radiation diagnostics help determine the stages of inflammatory bowel diseases helping in timely diagnosis, prescription and timely correction of therapy. The use of an adequate amount of diagnostic measures allows early detection of complications requiring immediate surgical treatment, such as fibrostenotic lesions, fistulae, and abscesses.The sources of the review were publications included in peer-reviewed databases E-library, PubMed, GoogleScholar, conference materials and non-indexed publications, including web pages of conferences and repositories of scientific data from universities.