Effectiveness of deep learning-based reconstruction for improvement of image quality and liver tumor detectability in the hepatobiliary phase of gadoxetic acid-enhanced magnetic resonance imaging

PURPOSE

To evaluate the effectiveness of deep learning-based reconstruction (DLR) in improving image quality and tumor detectability of isovoxel high-resolution breath-hold fat-suppressed T1-weighted imaging (HR-BH-FS-T1WI) in the hepatobiliary phase (HBP) of Gadoxetic acid-enhanced magnetic resonance imaging (Gd-EOB-MRI).

MATERIALS AND METHODS

This retrospective evaluated 42 patients with 98 liver tumors who underwent Gd-EOB-MRI between March 2023 and May 2023 using three techniques based on HBP imaging: isovoxel HR-BH-FS-T1WI reconstructed (1) with DLR (BH-DLR +) and (2) without DLR (BH-DLR -) and (3) HR-FS-T1WI scanned with a free-breathing technique using a navigator-echo-triggered technique and DLR (Navi-DLR +). The three techniques were qualitatively and quantitatively compared by the Friedman test and the Bonferroni post-hoc test. Tumor detectability was compared using the McNemar test.

RESULTS

BH-DLR + (3.85, average score of two radiologists) showed significantly better qualitative scores for image noise than BH-DLR - (2.84) and Navi-DLR + (3.37) (p < 0.0167), and Navi-DLR + showed significantly better scores than BH-DLR - (p < 0.0167). BH-DLR + (3.77) and BH-DLR - (3.77) showed significantly better qualitative scores for respiratory motion artifact than Navi-DLR + (2.75) (p < 0.0167), but there was no significant difference in scores between BH-DLR + and BH-DLR - (p > 0.0167). BH-DLR + (0.32) and Navi-DLR + (0.33) showed significantly higher lesion-to-nonlesion CR than BH-DLR - (0.29) (p < 0.0167), but there was no significant difference in lesion-to-nonlesion CR between BH-DLR + and Navi-DLR + (p > 0.0167). BH-DLR + (89.8%) showed significantly better tumor detectability than BH-DLR - (76.0%) and Navi-DLR + (77.6%) (p < 0.05).

CONCLUSION

The use of DLR for isovoxel HR-BH-FS-T1WI was effective in improving image quality and tumor detectability.

Improving automatic segmentation of liver tumor images using a deep learning model

Liver tumors are one of the most aggressive malignancies in the human body. Computer-aided technology and liver interventional surgery are effective in the prediction, identification and management of liver neoplasms. One of the important processes is to accurately grasp the morphological structure of the liver and liver blood vessels. However, accurate identification and segmentation of hepatic blood vessels in CT images poses a formidable challenge. Manually locating and segmenting liver vessels in CT images is time-consuming and impractical. There is an imperative clinical requirement for a precise and effective algorithm to segment liver vessels. In response to this demand, the current paper advocates a liver vessel segmentation approach that employs an enhanced 3D fully convolutional neural network V-Net. The network model improves the basic network structure according to the characteristics of liver vessels. First, a pyramidal convolution block is introduced between the encoder and decoder of the network to improve the network localization ability. Then, multi-resolution deep supervision is introduced in the network, resulting in more robust segmentation. Finally, by fusing feature maps of different resolutions, the overall segmentation result is predicted. Evaluation experiments on public datasets demonstrate that our improved scheme can increase the segmentation ability of existing network models for liver vessels. Compared with the existing work, the experimental outcomes demonstrate that the technique presented in this manuscript has attained superior performance on the Dice Coefficient index, which can promote the treatment of liver tumors.

Hepatic angiomyolipoma masquerading as abdominal pain: A case report

INTRODUCTION AND IMPORTANCE

Hepatic angiomyolipoma (HAML) is a rare liver tumor composed of blood vessels, smooth muscle, and fat cells. HAML occurs across a wide age range, with symptoms including abdominal discomfort, bloating, and weight loss. Diagnosis is challenging due to varied imaging appearances, but histopathological examination supplemented by immunohistochemical analysis, particularly using HMB-45, is definitive.

CASE PRESENTATION

A 33-year-old man presented with a two-year history of right upper quadrant abdominal pain, occasionally relieved with analgesics but worsening over the past month and a half. Examinations revealed a soft, non-distended abdomen with a palpable liver. Laboratory tests, including viral markers and tumor markers were normal. Contrast-enhanced CT revealed a well-defined oval mass in liver segment III with heterogeneous enhancement leading to provisional diagnosis of HAML. The patient underwent a successful en bloc excision with no intraoperative or postoperative complications.

CLINICAL DISCUSSION

Surgical resection is recommended for symptomatic cases or inconclusive biopsies, with stringent follow-up necessary due to the potential for recurrence and association with other malignancies.

CONCLUSION

HAML may present with prolonged nonspecific abdominal symptoms. CT imaging aids in diagnosing cases with abundant fatty tissue. En bloc tumor excision proves safe and effective in treating symptomatic presentations.

Integrated genotype-phenotype analysis of familial adenomatous polyposis-associated hepatocellular adenomas

Familial adenomatous polyposis (FAP) is an autosomal dominant syndrome caused by a germline mutation in the adenomatous polyposis coli (APC) gene, characterized by numerous colorectal adenomas. In addition, FAP patients may develop extraintestinal manifestations. Several cases of hepatocellular adenomas (HCA) detected accidentally in FAP patients have raised the so-far unsolved question of whether they represent a specific manifestation of FAP or a mere coincidence. To investigate the incidence of liver tumors in FAP patients, we analyzed our diagnostic database from 1991 to 2021. Among the 58 hepatic mass lesions identified, five HCAs occurring in three patients with FAP were identified, and comprehensive morphological, immunohistological, and molecular analysis employing targeted next-generation sequencing was conducted for characterization. The HCAs in this study showed no cytological or histological atypia. They displayed a diffuse, strong positivity for glutamine synthetase but no nuclear beta-catenin immunostaining. In two patients, the adenomas showed moderate immunoreactivity against serum amyloid A. Consistent with the diagnosis of FAP, molecular profiling revealed a pathogenic germline mutation of the APC gene in all analyzed adenomas as well as deleterious somatic second hits. All somatic mutations were localized between codons 1345 and 1577. No mutations were found in the catenin beta 1 gene. HCA in FAP patients can be a specific, although rare, neoplastic manifestation of this inborn disease and represents a distinct subgroup of HCAs. These benign tumors represent an important differential diagnosis for hepatic metastases in FAP patients and require adequate clinical and molecular (diagnostic) assessments for optimal patient guidance.

Thermal ablation for hepatic tumors in high-risk locations

Thermal ablative techniques such as radiofrequency and microwave ablation are minimally invasive and cost-effective approaches that are currently being adopted as alternatives to surgical resection for primary and metastatic liver malignancies. However, they are considered to be relatively contraindicated for tumors in high-risk locations due to technical difficulties and a perceived increased risk of perioperative complications. Several techniques, including artificial ascites, non-touch multibipolar ablation, and laparoscopically assisted ablation, can be used to improve the outcomes of ablation for high-risk tumors. This review aims to provide a comprehensive summary of the techniques currently used to improve thermal ablation outcomes for high-risk liver tumors.

Single-arm prospective study comparing ablation zone volume between time zero and 24 h after microwave ablation of liver tumors

PURPOSE

Percutaneous thermal ablation is an effective treatment for primary and metastatic liver tumors and is a recommended local therapy for early-stage hepatocellular carcinoma (HCC). Reported evidence shows an increase in the ablation zone volume over the first 24-h post-liver ablation. This report compares ablation zone volumes immediately at the completion (T = 0) of 26 microwave ablations of liver tumors to 24-h post-procedure (T = 24) volumes.

MATERIALS AND METHODS

20 patients, 13 (65%) males, underwent a total of 26 hepatic microwave ablations (MWA) under ultrasound guidance. Contrast-enhanced CT (CECT) or MRI was performed immediately and another CECT 24 h post operatively. Evaluation of the ablation zone and comparison of the two post-operative scans were done using BioTrace software. The expansion of ablation zones on post-op CECTs was matched point by point per direction. The distance between each 2 points was measured and grouped by distance. The incidence of each specific distance was then converted into a percentage, first for each case separately, then for all cases altogether. Data were tested by a matched paired one-sided t test.

RESULTS

The median lesion diameter was 1.5 cm (range 0.5-3.3) with 16 (62%) HCC cases and 9 hepatic metastases (4 neuroendocrine carcinoma, 4 colorectal carcinomas, 1 breast carcinoma, 1 pancreatic cancer). The data show a consistent volume expansion greater than 30% (p = 7.7e-5) 24-h post-ablation, where the median expansion is 57%. Distances between T = 0 and T = 24 equal to 3-7 mm occur in over 35% of the cases.

CONCLUSION

The ablation zone expansion at 24-h post-op was not uniform. The final ablation zone is difficult to predict at the time of the procedure. The awareness of the ablation zone expansion is important when treating near-critical structures, managing the heat sink effect, and preserving liver parenchyma.

Interferon therapy improves survival in patients with hepatitis B virus-related hepatocellular carcinoma after curative surgery: a meta-analysis

BACKGROUND AND AIM

A novel study found interferon enhanced antitumor activity of anti-PD-1-based immunotherapy and played a crucial role in improving efficacy on HCC, but the opposite results about the efficacy of interferon on HBV-related HCC were obtained from previous clinical studies and meta-analyses. Thus, this meta-analysis aimed to re-evaluate whether interferon could improve survival and reduce recurrence of patients with HBV-related HCC after curative surgery.

METHODS

MEDLINE/PubMed, Cochrane Library, EMBASE, Web of Science and CNKI were searched for eligible studies from inception to November 2022 and a meta-analysis was done.

RESULTS

10 trials with a total of 2062 subjects were screened. Interferon significantly improved 1-, 2-, 3- and 5-year OS and 1-, 2- and 3-year DFS, and reduced 2-, 3- and 5-year recurrence rates of patients with HBV-related HCC after curative surgery. However, interferon did not improve 8-year OS and 5-year DFS, did not reduce 1-year recurrence rate.

CONCLUSIONS

Interferon may significantly reduce recurrence and improve DFS of patients with HBV-related HCC after curative surgery, and finally improve the OS. However, the efficacy advantage may gradually weaken as time goes on. The clinical application of interferon combined with NAs recommended in this meta-analysis is needed to be further studied.

Successful diagnosis of a localized liver tumor via transjugular liver biopsy after ascites hampered a percutaneous approach

Liver biopsy is a crucial diagnostic procedure for liver-related conditions. However, it is challenging to use in cases with substantial ascites or coagulopathy. Transjugular liver biopsy has been introduced as an alternative method to overcome these limitations. Although transjugular liver biopsy is commonly used for diffuse liver diseases, its application to localized tumors has been limited. We report the case of a 66-year-old male initially diagnosed with unresectable pancreatic carcinoma, who later developed liver metastasis, peritoneal dissemination, and ascites. Treatment planning required tumor re-evaluation, but percutaneous liver biopsy was not viable because of the rapid accumulation of ascites and its presence on the liver surface. However, transjugular liver biopsy was a suitable alternative, given the proximity of the tumor to the right hepatic vein. The procedure, performed under fluoroscopic and ultrasound guidance, successfully obtained 4 specimens that were promptly diagnosed as liver metastases originating from pancreatic cancer. This case underscored the effectiveness of transjugular liver biopsy in situations where percutaneous biopsy is challenging because of conditions such as ascites. The combination of ultrasound guidance and rapid specimen assessment by pathology technicians can enhance diagnostic success rates. Transjugular liver biopsy is a valuable diagnostic tool in scenarios where percutaneous access poses significant difficulties.

USC-ENet: a high-efficiency model for the diagnosis of liver tumors combining B-mode ultrasound and clinical data

Purpose

Ultrasound image acquisition has the advantages of being low cost, rapid, and non-invasive, and it does not produce radiation. Currently, ultrasound is widely used in the diagnosis of liver tumors. However, owing to the complex presentation and diverse features of benign and malignant liver tumors, accurate diagnosis of liver tumors using ultrasound is difficult even for experienced radiologists. In recent years, artificial intelligence-assisted diagnosis has proven to provide effective support to radiologists. However, there is room for further improvement in the existing ultrasound artificial intelligence diagnostic model of liver tumor. First, the image diagnostic model may not fully consider relevant clinical data in the decision-making process. Second, owing to the difficulty in collecting biopsy pathology and physician-labeled ultrasound data of liver tumors, training datasets are usually small, and commonly used large neural networks tend to overfit on small datasets, which seriously affects the generalization of the model.

Methods

In this study, we propose a deep learning-assisted diagnosis model called USC-ENet, which integrates B-mode ultrasound features of liver tumors and clinical data of patients, and we design a small neural network specifically for small-scale medical images combined with an attention mechanism.

Results and conclusion

Real data from 542 patients with liver tumors (N = 2168 images) are used during model training and validation. Experiments show that USC-ENet can achieve a good classification effect (area under the curve = 0.956, sensitivity = 0.915, and specificity = 0.880) after small-scale data training, and it has certain interpretability, showing good potential for clinical adoption. In conclusion, our model provides not only a reliable second opinion for radiologists but also a reference for junior radiologists who lack clinical experience.

SAGES/AHPBA guidelines for the use of microwave and radiofrequency liver ablation for the surgical treatment of hepatocellular carcinoma or colorectal liver metastases less than 5 cm

BACKGROUND

Primary hepatocellular carcinoma (HCC) and colorectal liver metastases (CRLM) represent the liver's two most common malignant neoplasms. Liver-directed therapies such as ablation have become part of multidisciplinary therapies despite a paucity of data. Therefore, an expert panel was convened to develop evidence-based recommendations regarding the use of microwave ablation (MWA) and radiofrequency ablation (RFA) for HCC or CRLM less than 5 cm in diameter in patients ineligible for other therapies.

METHODS

A systematic review was conducted for six key questions (KQ) regarding MWA or RFA for solitary liver tumors in patients deemed poor candidates for first-line therapy. Subject experts used the GRADE methodology to formulate evidence-based recommendations and future research recommendations.

RESULTS

The panel addressed six KQs pertaining to MWA vs. RFA outcomes and laparoscopic vs. percutaneous MWA. The available evidence was poor quality and individual studies included both HCC and CRLM. Therefore, the six KQs were condensed into two, recognizing that these were two disparate tumor groups and this grouping was somewhat arbitrary. With this significant limitation, the panel suggested that in appropriately selected patients, either MWA or RFA can be safe and feasible. However, this recommendation must be implemented cautiously when simultaneously considering patients with two disparate tumor biologies. The limited data suggested that laparoscopic MWA of anatomically more difficult tumors has a compensatory higher morbidity profile compared to percutaneous MWA, while achieving similar overall 1-year survival. Thus, either approach can be appropriate depending on patient-specific factors (very low certainty of evidence).

CONCLUSION

Given the weak evidence, these guidelines provide modest guidance regarding liver ablative therapies for HCC and CRLM. Liver ablation is just one component of a multimodal approach and its use is currently limited to a highly selected population. The quality of the existing data is very low and therefore limits the strength of the guidelines.

High-security automatic path planning of radiofrequency ablation for liver tumors

BACKGROUND AND OBJECTIVE

Radiofrequency ablation (RFA) is an effective method for the treatment of liver tumors. Preoperative path planning, which plays a crucial role in RFA treatment, requires doctors to have significant experience and ability. Specifically, correct and highly active preoperative path planning should ensure the safety of the whole puncturing process, complete ablation of tumors and minimal damage to healthy tissues.

METHODS

In this paper, a high-security automatic multiple puncture path planning method for liver tumors is proposed, in which the optimization of the ablation number, puncture number, target positions and puncture point positions subject to comprehensive clinical constraints are studied. In particular, both the safety of the puncture path and the distribution of ablation ellipsoids are taken into consideration. The influence of each constraint on the safety of the whole puncturing process is discussed in detail. On this basis, the efficiency of the planning method is obviously improved by simplifying the computational data and optimized variables. In addition, the performance and adaptability of the proposed method to large and small tumors are compared and summarized.

RESULTS

The proposed method is evaluated on 10 liver tumors of various geometric characteristics from 7 cases. The test results show that the average path planning time and average ablation efficiency are 41.4 s and 60.19%, respectively. For tumors of different sizes, the planning results obtained from the proposed method have similar healthy tissue coverage. Through the clinical evaluation of doctors, the planning results meet the needs of RFA for liver tumors.

CONCLUSIONS

The proposed method can provide reasonable puncture paths in RFA planning, which is beneficial to ensure the safety and efficiency of liver tumor ablation.

A heuristic method for rapid and automatic radiofrequency ablation planning of liver tumors

PURPOSE

Preprocedural planning is a key step in radiofrequency ablation (RFA) treatment for liver tumors, which is a complex task with multiple constraints and relies heavily on the personal experience of interventional radiologists, and existing optimization-based automatic RFA planning methods are very time-consuming. In this paper, we aim to develop a heuristic RFA planning method to rapidly and automatically make a clinically acceptable RFA plan.

METHODS

First, the insertion direction is heuristically initialized based on tumor long axis. Then, the 3D RFA planning is divided into insertion path planning and ablation position planning, which are further simplified into 2D by projections along two orthogonal directions. Here, a heuristic algorithm based on regular arrangement and step-wise adjustment is proposed to implement the 2D planning tasks. Experiments are conducted on patients with liver tumors of different sizes and shapes from multicenter to evaluate the proposed method.

RESULTS

The proposed method automatically generated clinically acceptable RFA plans within 3 min for all cases in the test set and the clinical validation set. All RFA plans of our method achieve 100% treatment zone coverage without damaging the vital organs. Compared with the optimization-based method, the proposed method reduces the planning time by dozens of times while generating RFA plans with similar ablation efficiency.

CONCLUSION

The proposed method demonstrates a new way to rapidly and automatically generate clinically acceptable RFA plans with multiple clinical constraints. The plans of our method are consistent with the clinical actual plans on almost all cases, which demonstrates the effectiveness of the proposed method and can help reduce the burden on clinicians.

The Antitumor Impact of Combining Hepatic Artery Ligation With Copper Chelators for Liver Cancer

Background

Hepatocellular carcinoma (HCC) is one of the main cancer-related mortality worldwide. Thus, there is a constant search for improvement in treatment strategies to enhance the prognosis of this malignancy. The study aims to investigate the combined antitumor activity of ammonium tetrathiomolybdate (TM, copper chelator) combined with hepatic artery ligation (HAL) for liver cancer.

Methods

A total of 40 Sprague-Dawley (SD) rats bearing hepatic tumors were randomly divided into four groups: the control group without any treatment (control), HAL only (HAL), given TM by gavage (TM), and given TM combined with HAL (HAL + TM). The concentrations of serum copper were measured at the predetermined time points. Tumor growth rate, overall survival (OS), expression of hypoxia-inducible factor-1α (HIF-1α), vascular endothelial growth factor (VEGF), and microvessel density (MVD), as determined by immunohistochemical examination, were compared.

Results

HAL treatment transiently could elevate alanine transaminase (ALT) and aspartate transaminase (AST) but resumed to baseline within 1 week. Serum copper was significantly increased in tumor-bearing animals over time. The values of serum copper in the three treatment groups were significantly lower than those in the control group at different time points, with the lowest values observed in the TM group (P < .05). The average tumor size was 30.33 ± 2.58, 20.83 ± 2.93, 16.80 ± 3.84, and 10.88 ± 1.08 mm in the control, HAL, TM, and HAL + TM groups, respectively (HAL + TM vs other groups, all P < .05). In addition, the expression levels of HIF-1α, VEGF, and MVD were significantly lower in the HAL + TM group than those in the other groups (P < .05). The OS of rats in the combined groups was significantly prolonged combined to the other groups (P < .05), with survival time of 19.1 ± 0.64, 25.4 ± 1.24, 25.3 ± 1.78, and 29.9 ± 2.22 days in the control, HAL, TM, and HAL + TM groups, respectively.

Conclusion

These findings suggest that combined treatment with TM and HAL holds great potential for liver cancer treatment by reducing tumor hypoxia and angiogenesis. The observed results indicate that these combinations may offer a novel target and strategy for interventional therapy of liver cancer.

Efficacy and safety of laparoscopic liver resection for hepatocellular carcinoma in segment 8: a propensity score-matched two-center study

BACKGROUND

The safety and efficacy of laparoscopic liver resection (LLR) have been reported worldwide. However, those of LLR for tumors located in Couinaud's segment 8 are not sufficiently investigated.

METHODS

We retrospectively analyzed 108 patients who underwent liver resection for hepatocellular carcinoma (HCC) in segment 8 at Kobe University Hospital and Hyogo Cancer Center between January 2010 and December 2021. The patients were categorized in LLR and open liver resection (OLR) groups, and 1:1 propensity score matching (PSM) was performed to compare surgical outcomes between the groups.

RESULTS

Forty-seven and 61 patients underwent LLR and OLR, respectively. After PSM, each group contained 34 patients. There was no significant difference in operation time between the groups (331 min vs. 330 min, P = 0.844). Patients in the LLR group had significantly less blood loss (30 mL vs. 468 mL, P < 0.001) and shorter length of postoperative hospital stay (10 days vs. 12 days, P = 0.015) than those in the OLR group. There was no significant difference in the occurrence of postoperative complications between the groups (12% vs. 9%, P = 0.690). Further, the 1-year cumulative incidence of recurrence was not significantly different between the groups (16% vs. 19%, P = 0.734).

CONCLUSIONS

The surgical outcomes and short-term prognosis of LLR were similar or better than those of OLR. LLR could be an effective and safe procedure, even for lesions located in segment 8, which is considered a difficult anatomical location for LLR.

Small extrachromosomal circular DNA harboring targeted tumor suppressor gene mutations supports intratumor heterogeneity in mouse liver cancer induced by multiplexed CRISPR/Cas9

BACKGROUND

Primary liver cancer has significant intratumor genetic heterogeneity (IGH), which drives cancer evolution and prevents effective cancer treatment. CRISPR/Cas9-induced mouse liver cancer models can be used to elucidate how IGH is developed. However, as CRISPR/Cas9 could induce chromothripsis and extrachromosomal DNA in cells in addition to targeted mutations, we wondered whether this effect contributes to the development of IGH in CRISPR/Cas9-induced mouse liver cancer.

METHODS

CRISPR/Cas9-based targeted somatic multiplex-mutagenesis was used to target 34 tumor suppressor genes (TSGs) for induction of primary liver tumors in mice. Target site mutations in tumor cells were analyzed and compared between single-cell clones and their subclones, between different time points of cell proliferation, and between parental clones and single-cell clones derived from mouse subcutaneous allografts. Genomic instability and generation of extrachromosomal circular DNA (eccDNA) was explored as a potential mechanism underlying the oscillation of target site mutations in these liver tumor cells.

RESULTS

After efficiently inducing autochthonous liver tumors in mice within 30-60 days, analyses of CRISPR/Cas9-induced tumors and single-cell clones derived from tumor nodules revealed multiplexed and heterogeneous mutations at target sites. Many target sites frequently displayed more than two types of allelic variations with varying frequencies in single-cell clones, indicating increased copy number of these target sites. The types and frequencies of targeted TSG mutations continued to change at some target sites between single-cell clones and their subclones. Even the proliferation of a subclone in cell culture and in mouse subcutaneous graft altered the types and frequencies of targeted TSG mutations in the absence of continuing CRISPR/Cas9 genome editing, indicating a new source outside primary chromosomes for the development of IGH in these liver tumors. Karyotyping of tumor cells revealed genomic instability in these cells manifested by high levels of micronuclei and chromosomal aberrations including chromosomal fragments and chromosomal breaks. Sequencing analysis further demonstrated the generation of eccDNA harboring targeted TSG mutations in these tumor cells.

CONCLUSIONS

Small eccDNAs carrying TSG mutations may serve as an important source supporting intratumor heterogeneity and tumor evolution in mouse liver cancer induced by multiplexed CRISPR/Cas9.

Guide to use of ferumoxytol for hepatic vascular assessment as part of dual contrast MRI

Magnetic resonance imaging (MRI) assessment of hepatic vasculature can be challenging in the setting of liver disease and liver lesions. The widely used hepatobiliary contrast agent gadoxetate is an extracellular contrast agent that provides excellent soft tissue characterization but has limitations as a vascular contrast agent. Ferumoxytol is an iron oxide nanoparticle with superparamagnetic properties that can be used as blood pool contrast agent to provide dedicated vascular assessment. We provide a detailed protocol for evaluation of pediatric liver vasculature using ferumoxytol, after imaging of the parenchyma with gadoxetate. We provide multiple examples and discuss practical considerations when incorporating ferumoxytol into practice.

Novel breath-hold liver target stereotactic ablative radiotherapy using the intrafraction diaphragm registration of kilovoltage projection streaming image with digitally reconstructed radiography of the planning computed tomography

Stereotactic ablative radiotherapy (SABR) is an emerging treatment option for patients with primary or metastatic liver tumors, particularly for those who are not eligible for surgery or transplantation. SABR is a high-precision radiation therapy that delivers a high dose of radiation to the tumor while minimizing the dose to the surrounding healthy tissues. However, the accurate targeting of the tumor is a crucial aspect of liver SABR, which requires real-time imaging and tracking of the liver and tumor motion during treatment. One of the motion management strategies for liver SABR is the repeated breath-hold technique, which involves the patient holding their breath multiple times during treatment delivery to reduce the movement of the liver and other organs due to breathing. This technique helps to improve the accuracy of the treatment and reduce the radiation dose to the healthy liver. The current study proposes a novel approach for multiple breath-hold volumetric modulated arc therapy (VMAT) stereotactic ablative radiotherapy for liver tumors, which uses the intrafraction diaphragm registration in real time to improve the accuracy and precision of the treatment. The proposed approach is based on real-time comparison of the diaphragmatic surface location between the digitally reconstructed radiography (DRR) and intrafraction kilovoltage projection streaming images (kV-PSI) having the same beam angles. The image cross-correlation between the DRR and the intrafraction kV-PSI provides a measure of the similarity between the two images and can be used to identify and track the diaphragm position during VMAT delivery. The proposed methodology consists of several steps, including planning CT and treatment planning, reference image reconstruction, and patient positioning and immobilization. The proposed approach has the potential to improve the accuracy and precision of liver cancer VMAT SABR, thereby increasing the efficacy of the treatment and reducing the risk of radiation exposure to surrounding healthy tissues.

Application of a preoperative scoring system for small liver lesions in predicting small hepatocellular carcinoma

BACKGROUND

The treatment of small hepatocellular carcinoma (SHCC) is mainly based on surgical procedures, and early treatment can significantly improve the prognosis. How to improve the diagnostic rate of SHCC in patients with small liver lesions is a challenge faced by clinical doctors. Therefore, assessing the risk of SHCC in patients with small liver lesions and taking corresponding intervention measures are crucial for prolonging patient survival.

AIM

To evaluate the application value of a preoperative scoring system for small liver lesions established based on common clinical indicators in the clinical diagnosis of SHCC.

METHODS

The clinical data of 243 patients with small liver lesions who underwent surgery at our hospital from November 2014 to December 2022 were retrospectively analyzed. Multivariate logistic regression analysis was used to screen the variables with significant differences between the two groups of patients, and the weights of the main risk factors were assigned according to the screened values. A preoperative scoring system for small liver lesions was established. Finally, receiver operating characteristic curve (ROC) analysis was performed to evaluate its efficacy in the diagnosis of SHCC.

RESULTS

Logistic regression analysis showed that four variables, including viral hepatitis, cirrhosis, preoperative AFP level elevation, and imaging diagnosis of malignant tumor, were significantly associated with the occurrence of SHCC (P < 0.05). ROC analysis showed that the accuracy of the preoperative scoring system for small liver lesions in the diagnosis of malignant tumors was 0.864, and the cut-off value was 3.5. Based on the distribution of scores and ROC analysis results of the two groups of patients, the risk of developing SHCC in patients can be graded and analyzed.

CONCLUSION

The preoperative scoring system of small liver lesions has good efficacy for early diagnosis of benign and malignant small liver lesions.

Primary hepatic Burkitt lymphoma in a child and review of literature

BACKGROUND

Primary hepatic Burkitt lymphoma (PHBL) in children is an extremely rare hepatic malignancy with a dismal prognosis, unless it is detected and treated promptly. An 11-year-old child with abdominal pain was admitted to our hospital. No notable abnormalities were found during his physical examination or laboratory workup, but the abdominal computed tomography and magnetic resonance imaging both indicated a malignant hepatic mass measuring 9.2 × 7.1 × 7.5 cm in size. His postoperative pathology revealed an unexpected primary hepatic Burkitt lymphoma following a laparoscopic liver lobectomy. He then received rituximab and intense multi-agent chemotherapy as treatment. Despite post-chemotherapy bone marrow suppression, the patient eventually made a full recovery and had a good overall state.

CONCLUSION

In this study, we describe a rare case of pediatric primary hepatic Burkitt lymphoma and review the literature on clinical features, diagnosis, and treatment for primary hepatic Burkitt lymphoma in children. We stress that this diagnosis should be taken into account in the absence of other single hepatic lesions or primary tumors of hematological disorders, particularly when there is a normal AFP level.

[Diagnostic value of contrast-enhanced ultrasound in hepatic epithelioid hemangioendothelioma]

Objective: To investigate the features of contrast-enhanced ultrasound (CEUS) in hepatic epithelioid hemangioendothelioma (HEHE) in order to improve the preoperative diagnosis rate. Methods: CEUS images of 32 pathologically-proven cases of hepatic epithelioid hemangioendothelioma from January 2004 to August 2021 were collected. Lesions were analyzed to observe the features of enhancement mode, enhancement intensity, and distinct enhancement phases. Results: Among the 32 cases, one had a solitary lesion, 29 had multiple lesions, and two had diffuse-type lesions. Contrast-enhanced ultrasound revealed a total of 42 lesions in 32 cases. In terms of arterial phase enhancement, 18 lesions had overall enhancement, six lesions had uneven dendritic enhancement, 16 lesions had rim-like enhancement, and two lesions had just slight peripheral spot enhancement around the lesions. Among the three cases, there were multiple lesions that had overall enhancement and ring enhancement. In terms of the enhancement phase, 20 lesions showed "fast progression", 20 lesions showed "same progression", and two lesions showed "slow progression". During the late arterial or early portal venous phases with rapid washout, all lesions manifested as hypoechoic. With peaked enhanced intensity, 11 lesions had a lower enhancement intensity than the surrounding normal liver parenchyma; 11 lesions had the same enhancement degree as the surrounding normal liver parenchyma; and 20 lesions had a higher enhancement degree than the surrounding normal liver parenchyma. All 16 ring-enhancing lesions had marked hyperenhancement. In the typical enhancing lesions, four showed hyperenhancement, five showed low enhancement, and nine showed isoenhancement. In the dendrite-enhancing lesions, there were two isoenhancing and four hypoenhancing. Contrast-enhanced ultrasound delineated the boundaries of all lesions more clearly than two-dimensional ultrasound. Conclusion: Contrast-enhanced ultrasound has certain value in the diagnosis of hepatic epithelioid hemangioendothelioma.

Glycolysis, a new mechanism of oleuropein against liver tumor

BACKGROUND

Benign and malignant liver tumors are prevalent worldwide. However, there is no effective and comprehensive treatment option for many patients with malignant tumors. Thus, it is critical to prevent benign tumors from worsening, increasing the number of treatment options and effective medications against malignant liver tumors. Oleuropein is a natural and non-toxic product and inhibits tumor growth in various ways.

METHODS

We employed bioinformatics analysis and molecular docking to identify potential targets of oleuropein. Surface plasmon resonance (SPR) was used to determine the direct binding strength of the target and compounds. Essential functionalities of the targets were analyzed using gene interference approaches. Transcriptomic studies were performed to observe the global genomic alterations occurring inside cells. Changes in glycolytic metabolites and gene and protein expressions were also detected. The anti-tumor benefits of oleuropein in vivo were determined using a tumor-bearing mouse model.

RESULTS

Glucose-6-phosphate isomerase (GPI) was found to be a direct target of oleuropein. GPI discontinuation in liver tumor cells altered the expression of many genes, causing glycogenolysis. GPI interference was associated with PYGM and PFKFB4 inhibitors to inhibit glycolysis in liver tumors. Oleuropein inhibited glycolysis and showed good anti-tumor activity in vivo without adverse side effects.

CONCLUSIONS

GPI is a crucial enzyme in glycolysis and the immediate target of oleuropein. GPI expression inside tumor cells affects different physiological functions and signal transduction. Oleuropein has depicted anti-tumor action in vivo without harmful side effects. Moreover, it can control tumor glycolysis through GPI.

[Multilocular hepatic masses due to Enterobius vermicularis]

Infestation with Enterobius vermicularis involving extraintestinal manifestation is rare. In a 50-year-old man with lower abdominal pain, computed tomography led to a suspected diagnosis of sigmoid carcinoma with liver metastasis. After ruling out colon cancer by endoscopy, laparoscopic resection of one suspected tumor nodule in the liver was performed. Histopathological examination revealed parasitic granulomas containing pinworms. Hematogenous migration was postulated in the setting of phlebitis with thrombosis of the inferior mesenteric vein.

Leverage prior texture information in deep learning-based liver tumor segmentation: A plug-and-play Texture-Based Auto Pseudo Label module

Segmenting the liver and tumor regions using CT scans is crucial for the subsequent treatment in clinical practice and radiotherapy. Recently, liver and tumor segmentation techniques based on U-Net have gained popularity. However, there are numerous varieties of liver tumors, and they differ greatly in terms of their shapes and textures. It is unreasonable to regard all liver tumors as one class for learning. Meanwhile, texture information is crucial for the identification of liver tumors. We propose a plug-and-play Texture-based Auto Pseudo Label (TAPL) module to take use of the texture information of tumors and enable the neural network actively learn the texture differences between various tumors to increase the segmentation accuracy, especially for small tumors. The TPAL module consists of two parts, texture enhancement and texture-based pseudo label generator. To highlight the regions where the texture varies significantly, we enhance the textured areas of the CT image. Based on their texture information, tumors are automatically divided into several classes by the texture-based pseudo label generator. The multi-class tumors produced by the neural network during the prediction step are combined into a single tumor label, which is then used as the outcome of the segmentation. Experiments on clinical dataset and public dataset Lits2017 show that the proposed algorithm outperforms single liver tumor label segmentation methods and is more friendly to small tumors.

Ablation for Benign Liver Tumors: Current Concepts and Limitations

Percutaneous ablation under imaging guidance is a curative treatment that can induce complete tumor necrosis with advantages of minimal invasiveness and a low risk of complications. Thermal ablation, which includes radiofrequency ablation and microwave ablation, is a representative technique that has sufficient antitumor effects in cases of hepatocellular carcinoma with ≤3 lesions measuring ≤3 cm and preserved liver function. The short- and long-term outcomes of patients are comparable with those achieved with surgical resection. Despite their nonmalignant nature, some benign liver tumors require treatment for symptoms caused by the presence of the tumor and/or continuous enlargement. Ablation may be the treatment of choice because it has lower burden on patients than surgical treatment. This review describes the recent concepts, progress, and limitations of ablation-based treatment for benign liver tumors.

Improving Outcome of Selected Patients With Non-Resectable Hepatic Metastases From Colorectal Cancer With Liver Transplantation: A Prospective Parallel Trial (COLT trial)

BACKGROUND

Patients with unresectable Colorectal Liver Metastases (CLM) receiving palliative chemotherapy have a 5-year overall survival (OS) of less than 30%. Liver transplantation (LT) can improve OS up to 60%-83% (SECA-I and SECA-II trials). The aim of the study is to assess the efficacy of LT in liver-only metastatic CRC compared with a matched cohort of patients included in a phase III trial on triplet chemotherapy + antiEGFR.

PATIENTS AND METHODS

The COLT trial is an investigator-driven, multicenter, non-randomized, open-label, controlled, prospective, parallel trial (ClinicalTrials.gov NCT03803436). Hyperselected patients with liver-limited unresectable CLM, RAS and BRAF wild-type and curatively removed primary colon cancer are included. The observed post-transplant outcomes will be prospectively compared 1:5 with those obtained in a matched cohort from the TRIPLETE trial (NCT03231722).

RESULTS

Primary endpoint is to compare the 3 and 5-years OS of patients enrolled in the COLT trial with COLT-eligible population enrolled in the TRIPLETE trial. An expected gain in OS of 40% at 5-years is predicted for the COLT population (the expected OS at 5-years in COLT vs. TRIPLETE is 70% vs. 30%). Secondary endpoints are to compare the 5-years disease-free survival and to assess the safety of LT (Dindo-Clavien Classification and the Comprehensive Complication Index).

CONCLUSION

LT offers the longest OS reported in selected patients with CLM. Improving the selection strategies can give patients a 5-year OS similar to other indications for LT and a better outcome than those undergoing chemotherapy alone.

The Liver Tumor Segmentation Benchmark (LiTS)

In this work, we report the set-up and results of the Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI) 2017 and the International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2017 and 2018. The image dataset is diverse and contains primary and secondary tumors with varied sizes and appearances with various lesion-to-background levels (hyper-/hypo-dense), created in collaboration with seven hospitals and research institutions. Seventy-five submitted liver and liver tumor segmentation algorithms were trained on a set of 131 computed tomography (CT) volumes and were tested on 70 unseen test images acquired from different patients. We found that not a single algorithm performed best for both liver and liver tumors in the three events. The best liver segmentation algorithm achieved a Dice score of 0.963, whereas, for tumor segmentation, the best algorithms achieved Dices scores of 0.674 (ISBI 2017), 0.702 (MICCAI 2017), and 0.739 (MICCAI 2018). Retrospectively, we performed additional analysis on liver tumor detection and revealed that not all top-performing segmentation algorithms worked well for tumor detection. The best liver tumor detection method achieved a lesion-wise recall of 0.458 (ISBI 2017), 0.515 (MICCAI 2017), and 0.554 (MICCAI 2018), indicating the need for further research. LiTS remains an active benchmark and resource for research, e.g., contributing the liver-related segmentation tasks in http://medicaldecathlon.com/. In addition, both data and online evaluation are accessible via https://competitions.codalab.org/competitions/17094.

Primary hepatic myopericytoma coexisting with multiple cystic hepatic lesions: a case report

BACKGROUND

Hepatic myopericytoma (MPC) is an extremely rare pathological entity in the liver. Conversely, cystic hepatic lesions are a group of heterogeneous lesions encountered commonly in daily practice. Here, we report a unique case of the coexistence of primary hepatic MPC and multiple cystic hepatic lesions along with our perceptions on its diagnosis and treatment.

CASE PRESENTATION

A 56-year-old female patient was found to have a left liver mass during a routine physical examination. Computer tomography (CT) and magnetic resonance imaging (MRI) confirmed the existence of a left hepatic neoplasm along with multiple hepatic cysts but could not exclude the possible malignant nature of the neoplasm. Computer tomography (CT) also identified an enlarged mediastinal lymph node with a maximum diameter of 4.3 cm, which further underwent core needle biopsy under CT guidance. A histopathological examination was performed to rule out malignancy. Afterwards, the patient underwent left hemihepatectomy to resect a solid tumor of 5.5 cm × 5 cm × 4.7 cm with multiple cystic lesions which were histopathologically examined to establish the diagnosis of myopericytoma with hepatic cysts. Postoperatively, the patient recovered from the surgery quickly without significant adverse events and was not found to have a reoccurrence of the primary pathological entity.

CONCLUSIONS

This is the first reported case of a patient with the co-existence of primary hepatic myopericytoma and multiple cystic hepatic lesions undergoing surgical treatment with eventual recovery.

Estimation of post-therapeutic liver reserve capacity using 99mTc-GSA scintigraphy prior to carbon-ion radiotherapy for liver tumors

BACKGROUND

There is currently no established imaging method for assessing liver reserve capacity prior to carbon-ion radiotherapy (CIRT) for liver tumors. In order to perform safe CIRT, it is essential to estimate the post-therapeutic residual reserve capacity of the liver.

PURPOSE

To evaluate the ability of pre-treatment 99mTc-galactosyl human serum albumin (99mTc-GSA) scintigraphy to accurately estimate the residual liver reserve capacity in patients treated with CIRT for liver tumors.

MATERIALS AND METHODS

This retrospective study evaluated patients who were performed CIRT for liver tumors between December 2018 and September 2020 and underwent 99mTc-GSA scintigraphy before and 3 months after CIRT, and gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced MRI within 1 month before CIRT were evaluated. The maximal removal rate of 99mTc-GSA (GSA-Rmax) was analyzed for the evaluation of pre-treatment liver reserve capacity. Then, the GSA-Rmax of the estimated residual liver (GSA-RL) was calculated using liver SPECT images fused with the Gd-EOB-DTPA-enhanced MRI. GSA-RL before CIRT and GSA-Rmax at 3 months after CIRT were compared using non-parametric Wilcoxon signed-rank test and linear regression analysis.

RESULTS

Overall, 50 patients were included (mean age ± standard deviation, 73 years ± 11; range, 29-89 years, 35 men). The median GSA-RL was 0.393 [range, 0.057-0.729] mg/min, and the median GSA-Rmax after CIRT was 0.369 [range, 0.037-0.780] mg/min (P = .40). The linear regression equation representing the relationship between the GSA-RL and GSA-Rmax after CIRT was y = 0.05 + 0.84x (R2 = 0.67, P < .0001). There was a linear relationship between the estimated and actual post-treatment values for all patients, as well as in the group with impaired liver reserve capacity (y = - 0.02 + 1.09x (R2 = 0.62, P = .0005)).

CONCLUSIONS

99mTc-GSA scintigraphy has potential clinical utility for estimating the residual liver reserve capacity in patients undergoing carbon-ion radiotherapy for liver tumors.

TRIAL REGISTRATION

UMIN000038328, https://center6.umin.ac.jp/cgi-open-bin/ctr/ctr_view.cgi?recptno=R000043545 .

Multitargeting Strategy Using Tetrathiomolybdate and Lenvatinib: Maximizing Antiangiogenesis Activity in a Preclinical Liver Cancer Model

PURPOSE

The study aims to investigate the suppressing tumor-promoting effects via multi-anti-angiogenesis activity of the copper chelator (ammonium tetrathiomolybdate, TM) combined with lenvatinib for hepatocellular carcinoma.

METHODS

A total of 55 C57 mice were injected subcutaneously with Hepa1-6 hepatoma cell suspensions into the right posterior thigh. After 7 days, the subcutaneous tumors were formed, and the mice were randomly divided into five groups: TM (G1), Lenvatinib (G2), TM+Lenvatinib (G3), Control (G4), and Copper (II) Gluconate (G5). The copper concentrations in serum and tumors were measured at the predetermined time points. After 14 days of treatments, tumor weight and volumes were analyzed, histology was observed, and the expressions of vascular endothelial growth factor (VEGF) and microvessel density (MVD) in tumor tissues were measured by immunohistochemistry.

RESULTS

The median concentration of copper in serum was 401.70, 469.40, and 665.35 μg/L in normal mice, in mice 7 days after implantation, and in the control group, respectively. The intratumoral copper concentrations were higher in G4 mice than in mice 7 days after implantation (P < 0.05). The serum concentration of copper was higher in G5 than all the other groups (P < 0.05; (G1, G2, and G3) vs. G4, P < 0.05; G1 vs. G2, P = 0.013; G2 vs. G3, P = 0.018; G1 vs. G3, P = 0.903. The intratumoral copper concentrations were 608.40, 980.00, 539.31, and 2938.90 μg/L in G1, G2, G3, and G5, respectively. The average tumor weight was 0.55, 0.44, 0.08, 1.37, and 3.11 in G1, G2, G3, G4, and G5, respectively. G5 vs. other groups, P < 0.05; (G1, G2, and G3) vs. G4, P < 0.05; G1 vs. G3, P < 0.05; G2 vs. G3, P < 0.05; G1 vs. G2, P > 0.05. Furthermore, the expression levels of VEGF were significantly lower in G1, G2, and G3 than in G4 and G5 (P < 0.05). A similar trend was observed for MVD in the five groups, but no significant difference was detected in G1 and G2.

CONCLUSION

The study showed a significant positive correlation between tumor load and copper. Copper promotes tumor progression, but copper chelating suppresses tumor growth. The combination of TM with lenvatinib reduces tumor angiogenesis and improves the effect of antitumor treatment. These findings underlie the clinical application of combination therapy.

Can living donor liver transplantation provide similar outcomes to deceased-donor liver transplantation for hepatocellular carcinoma? A systematic review and meta-analysis

BACKGROUND AND AIM

A potential solution to the deceased organ shortage is to include live organ donations and to identify patients with lower rates of HCC recurrence to fairly allocate liver grafts. Our aims were to detect the long-term outcomes of LDLT versus DDLT for HCC and predictors of recurrence after transplantation.

METHODS

PubMed, Scopus, Web of Science, Cochrane library were searched for eligible studies from inception to July 2021 and a systematic review and meta-analysis were done.

RESULTS

35 studies with a total of 7822 patients were included. The 1-, 3-, 4 year-OS showed trivial improvement for LDLT recipients. However, the two modalities had similar 5-, 6- and 10-year OS. A significant improvement in the ITT-OS was observed for LDLT recipients. Regarding the DFS and recurrence after transplantation, no significant difference was observed between LDLT and DDLT. In addition to that, the pooled hazard ratio of the included studies showed that Milan criteria, level of AFP, presence of vascular invasion, tumor differentiation were significant predictors of recurrence.

CONCLUSION

The cancer biology (not the graft type) is the most important determinant of recurrence and survival after LT. However, LDLT provided much better survival benefits to HCC patients especially in regions that suffer from low deceased organ availability.

A study of the clinical profile, predictors, prognostic features, and survival of patients with hepatocellular carcinoma having macroscopic portal vein tumor thrombosis

BACKGROUND

Macroscopic portal vein tumor thrombosis (PVTT) is considered a negative prognostic factor in hepatocellular carcinoma (HCC) patients. There is divergent opinion regarding management of these patients worldwide. We aimed to evaluate the clinical profile, predictors, prognostic features, and survival of patients of HCC with PVTT.

METHODS

Treatment-naïve HCC patients with and without PVTT were analyzed retrospectively using a prospectively accrued dataset. Patients with PVTT were further divided as per treatment groups for survival analysis.

RESULTS

Of 508 patients, 46.1% had radiological evidence of PVTT at presentation. On logistic regression, serum albumin (odds ratio [OR]=0.65, 95% confidence interval [CI]= 0.44-0.96; p= 0.031); international normalized ratio (OR = 3.78,95% CI = 1.42-10.00; p=0.008); alpha-feto protein >400 ng/mL (OR=3.58, 95%CI = 2.00-6.40; p <0.001); size of largest tumor nodule >5 cm (OR =6.37, 95%CI =2.03-19.99; p =0.002); and male gender (OR =1.84, 95%CI = 1.01-3.33; p = 0.045) were independent predictors for PVTT. Patients with PVTT amenable to aggressive therapies had significantly better median overall survival (in months) as compared to those receiving sorafenib or best supportive care only (13.1, 3.9, and 1.8 respectively, p<0.0001). Treatment modality received (p<0.001) and extrahepatic metastasis (p=0.006) were independent predictors of mortality in these patients.

CONCLUSION

Size of largest tumor nodule >5 cm and alpha-fetoprotein >400 ng/mL are strongly associated with the presence of PVTT in patients with HCC. A multidisciplinary approach may identify a subgroup of patients who can be offered aggressive therapies like surgery and/or locoregional therapy with significant survival benefit.

Prevention of major biliary complications by fusion imaging for thermal ablation of malignant liver tumors adjacent to the bile ducts: a preliminary comparative study

PURPOSE

Ultrasound (US)-guided thermal ablation (TA) may cause major biliary complications, particularly in patients with malignant liver tumors (MLTs) adjacent to the bile ducts. Fusion imaging (FI), is postulated to reduce complication rate; however, there is a lack of clinical data to support this theory. Thus, the aim of our study was to evaluate the safety and efficacy of FI for TA of MLTs proximal to the bile ducts.

METHODS

A retrospective single-center review was conducted on a total of 289 patients with 316 MLTs adjacent to the bile ducts. The patients were divided into two groups based on whether FI was used in the ablation procedures. The choice of the FI-assisted procedure always depends on different operation periods and whether registrations will succeed. The baseline demographics and outcomes of these patients were compared. The efficacy was determined at the 1-month follow-up using contrast-enhanced computed tomography/magnetic resonance. Biliary complications and local tumor progression were subsequently followed-up every 3-6 months. The last follow-up visit was before August 30, 2019.

RESULTS

Among the included tumors, the incidence rate of major biliary complications after ablation in the FI group was 1.6%, which was significantly lower than that in the non-FI group (7.9%, p = 0.005). There was no significant difference in the efficacy rates of the techniques [99.5% (185/186) versus 98.4% (123/125), p = 0.56] or local progression rates [3.8% (7/185) versus 5.7% (7/123), p = 0.61] between the FI and non-FI groups.

CONCLUSION

FI for US-guided TA could be a noninvasive means to decrease major biliary complications. Trial registration number and date of registration: retrospectively registered.

Minimally invasive versus open hepatectomy for the resection of colorectal liver metastases: a systematic review and meta-analysis

BACKGROUND

While surgical resection has a demonstrated utility for patients with colorectal liver metastases (CRLM), it is unclear whether minimally invasive surgery (MIS) or an open approach should be used. This review sought to assess the efficacy and safety of MIS versus open hepatectomy for isolated, resectable CRLM when performed separately from (Key Question (KQ) 1) or simultaneously with (KQ2) the resection of the primary tumor.

METHODS

PubMed, Embase, Google Scholar, Cochrane CENTRAL, International Clinical Trials Registry Platform (ICTRP), and ClinicalTrials.gov databases were searched to identify both randomized controlled trials (RCTs) and non-randomized comparative studies published during January 2000-September 2020. Two independent reviewers screened literature for eligibility, extracted data from included studies, and assessed internal validity using the Cochrane Risk of Bias 2.0 Tool and the Newcastle-Ottawa Scale. A random-effects meta-analysis was performed using risk ratios (RR) and mean differences (MD).

RESULTS

From 2304 publications, 35 studies were included for meta-analysis. For staged resections, three RCTs and 20 observational studies were included. Data from RCTs indicated MIS having similar disease-free survival (DFS) at 1-year (RR 1.03, 95%CI 0.70-1.50), overall survival (OS) at 5-years (RR 1.04, 95%CI 0.84-1.28), fewer complications of Clavien-Dindo Grade III (RR 0.62, 95%CI 0.38-1.00), and shorter hospital length of stay (LOS) (MD -6.6 days, 95%CI -10.2, -3.0). For simultaneous resections, 12 observational studies were included. There was no evidence of a difference between MIS and the open group for DFS-1-year, OS-5-year, complications, R0 resections, blood transfusions, along with lower blood loss (MD -177.35 mL, 95%CI -273.17, -81.53) and shorter LOS (MD -3.0 days, 95%CI -3.82, -2.17).

CONCLUSIONS

Current evidence regarding the optimal approach for CRLM resection demonstrates similar oncologic outcomes between MIS and open techniques, however MIS hepatectomy had a shorter LOS, lower blood loss and complication rate, for both staged and simultaneous resections.

Suitability of Laparoscopic Liver Resection of Segment VII: a Retrospective Two-Center Study

BACKGROUND

Resecting liver tumors located in Couinaud's segment VII is challenging; the efficacy and safety of laparoscopic liver resection for segment VII lesions compared to open liver resection remain unclear.

METHODS

Medical records of 84 patients who underwent liver resection of segment VII at Kobe University Hospital and Hyogo Cancer Center between 2010 and 2021 were retrospectively analyzed. Surgical outcomes were compared between laparoscopic liver resection and open liver resection groups using propensity matching analysis.

RESULTS

Thirty-one and 53 patients underwent laparoscopic liver resection and open liver resection, respectively. After propensity matching, 29 patients were included in each group. The laparoscopic liver resection group had a significantly longer operation time (407 vs. 305 min, P = 0.002), lower blood loss (100 vs. 230 mL, P = 0.004), and higher postoperative alanine aminotransferase levels (436 vs. 252 IU/L, P = 0.008) than the open liver resection group. In patients with liver cirrhosis, the proportion of patients with postoperative liver-specific complications was higher in the laparoscopic liver resection group than in the open liver resection group (57% vs 11%, P = 0.049), although there was no significant difference in postoperative liver-specific complication rates between the groups in patients without liver cirrhosis.

CONCLUSIONS

For liver resection of segment VII, laparoscopic liver resection led to higher postoperative liver damage than open liver resection. Open liver resection may be better for patients with liver cirrhosis to avoid postoperative liver-specific complications. Laparoscopic liver resection could be an acceptable procedure for patients without liver cirrhosis, with some merits such as less blood loss.

Two Cases of Hepatoblastoma in Adults

Adult hepatoblastoma is a rare tumor whose etiology and mechanisms of development are still incompletely understood. Imaging and biological tests such as AFP and liver enzymes are non-specific. Histologically, there are 2 histological variants: pure epithelial with 5 types (pure fetal, embryonal, small cell undifferentiated, cholangioblastic, and macrotrabecular), a mixed epithelial and a mesenchymal variant with or without a teratoid contingent. The main differential diagnosis concerns hepatocellular carcinoma. The treatment of hepatoblastoma in adults is not yet standardized and surgery remains the mainstay of treatment. In this report we aim to describe the clinical, pathological, and immunohistochemical features of this rare entity in adult patients and discuss the elements allowing its distinction from hepatocellular carcinoma (HCC).

Hepatic Deletion of X-box Binding Protein 1 in Farnesoid X Receptor Null Mice Leads to Enhanced Liver Injury

Farnesoid X receptor (FXR) regulates bile acid metabolism, and FXR null (Fxr^-/-) mice have elevated bile acid levels and progressive liver injury. The inositol-requiring enzyme 1α (IRE1α)/X-box binding protein 1 (XBP1) pathway is a protective unfolded protein response (UPR) pathway activated in response to ER stress. Here, we sought to determine the role of the IRE1α/XBP1 pathway in hepatic bile acid toxicity using the Fxr^-/- mouse model. Western blotting and qPCR analysis demonstrated that hepatic XBP1 and other UPR pathways were activated in 24-week-old Fxr^-/- compared to 10-week-old Fxr^-/- mice, but not in WT mice. To further determine the role of the liver XBP1 activation in older Fxr^-/- mice, we generated mice with whole-body FXR and liver-specific XBP1 double knockout (DKO, Fxr^-/-Xbp1^LKO) and Fxr^-/-Xbp1^fl/fl single knockout (SKO) mice and characterized the role of hepatic XBP1 in cholestatic liver injury. Histologic staining demonstrated increased liver injury and fibrosis in DKO compared to SKO mice. RNA-seq revealed increased gene expression in apoptosis, inflammation, and cell proliferation pathways in DKO mice. The proapoptotic C/EBP-homologous protein (CHOP) pathway and cell cycle marker Cyclin D1 were also activated in DKO mice. Furthermore, we found total hepatic bile acid levels were similar between the two genotypes. At age 60 weeks, all DKO mice and no SKO mice spontaneously developed liver tumors. In conclusion, the hepatic XBP1 pathway is activated in older Fxr^-/- mice and has a protective role. The potential interaction between XBP1 and FXR signaling may be important in modulating the hepatocellular cholestatic stress responses.

Safety of CT-Guided Microwave Ablation of Subcardiac Liver Tumors

PURPOSE

The purpose of this study was to investigate the safety of CT-guided microwave ablation (MWA) of subcardiac hepatic tumors.

MATERIALS AND METHODS

This retrospective study included 19 patients (11 males and 8 females, age: 64.0 years (IQR: 58.3, 71.0) who underwent CT-guided MWA of 22 subcardiac tumors from January 2016 through December 2020. The subcardiac tumors consisted of 6 hepatocellular carcinomas and 16 metastases. Hydrodissection or other thermal protection technique was not used during the ablation. Subcardiac ablation was defined as the ablation zone extended ≤ 0.5 cm from myocardium or coronary artery. The safety of MWA of subcardiac tumors was evaluated based on procedural and post-procedural complications and intra-procedural ECG changes. Local tumor progression (LTP) was also analyzed and correlated with tumor and ablation zone sizes.

RESULTS

The primary efficacy rate was 100%. The median follow-up was 20.5 months (IQR: 6.0, 29.8). There was no 30-day mortality. One grade 3 complication occurred (severe shoulder and chest pain), and there were 19 events of grade 1 or 2 complications. No instances of cardiac complications or significant procedural ECG changes were observed. There were 22 events of grade 1 and 2 laboratory toxicity and 1 event of grade 3 elevated bilirubin. The LTP was 13.6% at 1 year and 22.7% at 2 years. There was no significant correlation between LTP and tumor or ablation zone sizes.

CONCLUSION

CT-guided MWA of subcardiac hepatic tumors is safe, and MWA should be considered as an option for managing subcardiac tumors. LTP rates for MWA of subcardiac tumors may be inferior to ablation of tumors in common location.

LEVEL OF EVIDENCE III

Cohort Study.

Integration of a dedicated management protocol in the care of pediatric liver cancer: From specialized providers to complication reduction

INTRODUCTION

Up to a third of children undergoing partial hepatectomy for primary hepatic malignancies experience at least one perioperative complication, with a presumed deleterious effect on both short- and long-term outcomes. We implemented a multidisciplinary treatment protocol in the management of these patients in order to improve complication rates following partial hepatectomy.

METHODS

A retrospective chart review was completed for all patients < 18 years of age who underwent liver resection at our institution between 2002 and 2019 for primary hepatic cancer. Demographic, intraoperative, postoperative, pathologic, and outcome data were analyzed for perioperative complications using the CLASSIC and Clavien-Dindo (CD) scales, event-free survival (EFS) and overall survival (OS).

RESULTS

A total of 73 patients were included in the analysis with 33 prior-to and 40 after dedicated provider protocol implementation. Perioperative complication rates decreased from 52% to 20% (p = 0.005) with major complications going from 18% to 10% (p = 0.31). On multivariable logistic regression, protocol implementation was associated with a reduction in any (OR 0.29 [95% CI 0.09 - 0.89]) but not major complications. On multivariate cox models, post protocol implementation was associated with improved event free survival (EFS) (HR 0.19 (0.036 - 0.195). Among patients with a diagnosis of hepatoblastoma (n = 62), the occurrence of a major perioperative complication was associated with a worse EFS (HR=5.45, p = 0.03) on multivariate analysis, however this did not translate into an impact on overall survival.

CONCLUSIONS

Our results demonstrate that, for children with primary liver malignancies, a dedication of patients to high-volume surgeons can improve rates of complications of liver resections and may improve the oncological outcome of hepatoblastoma.

A comparative study on computational models of multi-electrode radiofrequency ablation of large liver tumors

PURPOSE

Thermal ablation of liver tumors has emerged as a first-line curative treatment for single small tumors (diameter < 2.5 cm) due to similar overall survival rates as surgical resection. Moreover, it is far less invasive, has lower complication rates, a superior cost-effectiveness, and an extremely low treatment-associated mortality. However, in many cases, complete tumor coverage cannot be achieved only with a single electrode and several electrodes are used to create overlapping ablations. Multi-electrode planning is a challenging 3D task with many contradictive constraints to consider, a dimensionality difficult to assess even for experts. It requires extremely long planning time since it is mostly performed mentally by clinicians looking at 2D CT views. An accurate and reliable prediction of the ablation zone would help to turn thermal ablation into a first-line curative treatment also for large liver tumors treated with multiple electrodes. In order to determine the level of model simplification that can be acceptable, we compared three computational models, a simple spherical model, a biophysics-based model and an Eikonal model.

METHODS

RF ablation electrodes were virtually placed at a desired position in the patient pre-operative CT image and the models predicted the ablation zone generated by multiple electrodes. The last two models are patient-specific. In these cases, hepatic structures were automatically segmented from the pre-operative CT images to predict a patient-specific ablation zone.

RESULTS

The three models were used to simulate multiple electrode ablations on 12 large tumors from 11 patients for which the procedure information was available. Biophysics-based simulations approximate better the post-operative ablation zone in term of Hausdorff distance, Dice Similarity Coefficient, radius, and volume compared to two other methods. It also predicts better the coverage percentage and thus the tumor ablation margin.

CONCLUSION

The results obtained with the biophysics-based model indicate that it could improve ablation planning by accurately predicting the ablation zone, avoiding over or under-treatment. This is particularly beneficial for multi-electrode radiofrequency ablation of larger liver tumors where the planning phase is particularly challenging.

The study of direct and indirect effects of radiofrequency ablation on tumor microenvironment in liver tumor animal model

Background

Direct and indirect effects of radiofrequency ablation (RFA) on tumor microenvironment of the liver tumor have been noted, which was reported to be related to a variety of tyrosine protein kinase or cytokinetic pathway, but have not been thoroughly investigated and conclusive.

Purpose

To elucidate direct and indirect effects of RFA on tumor microenvironment in the liver tumor model, and to explore the role of the specific inhibitor in tumor growth by targeting the key pathway of RFA.

Materials and methods

One hundred and ten mice with H22 liver tumor were used in animal experiments. Eighty-four mice were randomized into three groups: control, direct RFA and indirect RFA (a block slide was inside the middle of the tumor). The growth rate of the residual tumor after RFA was calculated (n = 8 each group) and the pathologic changes at different time points (6 h, 24 h, 72 h and 7d after RFA) were evaluated (n = 5 in each subgroup). After semi-quantitative analysis of the pathological staining, the most significant marker after RFA was selected. Then, the specific inhibitor (PHA) was applied with RFA and the tumor growth and pathological changes were evaluated and compared with RFA alone. The Kruskal-Wallis test was used for evaluating the significance of different treatments in the pathological positive rate of specific markers in tumor. The two-way analysis of variance was used to determine the significance of treatment in tumor growth or body weight.

Results

The growth rate of the residual tumor in the direct RFA group was faster than the indirect RFA group (P = 0.026). The pathological analysis showed the expression of HSP70 (73 ± 13% vs 27 ± 9% at 24 h, P < 0.001), SMA (70 ± 18% vs 18 ± 7% at 6 h, P < 0.001) and Ki-67 (51 ± 11% vs 33 ± 14% at 7d, P < 0.001) in the direct RFA group was higher than those in the indirect RFA group after RFA. On the other hand, the expression of c-Met (38 ± 11% vs 28 ± 9% at 24 h, P = 0.01), IL-6 (41 ± 10% vs 25 ± 9% at 24 h, P < 0.001) and HIF-α (48 ± 10% vs 28 ± 8% at 24 h, P < 0.001) in the indirect RFA group was higher than those in the direct RFA group. And the expression of c-Met increased mostly in both direct and indirect RFA group compared to the baseline (53 and 65% at 72 h). Then the specific inhibitor of c-Met-PHA was applied with RFA. The growth rate of the tumor was significantly slower in the RFA + PHA group than the RFA alone group (1112.9 ± 465.6 mm³ vs 2162.7 ± 911.1 mm³ at day 16, P = 0.02).

Conclusion

Direct and indirect effects of RFA on tumor microenvironment changed at different time points and resulted in increased residual tumor growth in the animal model. It can be potentially neutralized with specific inhibitor of related pathways, such as tyrosine-protein kinase c-Met.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-022-09730-x.

Classification of hepatic cavernous hemangioma or hepatocellular carcinoma using a convolutional neural network model

BACKGROUND

Computed tomography (CT) is a common imaging technique for diagnosis of liver tumors. However, the intensity similarity on non-contrast CT images is small, making it difficult for radiologists to visually identify hepatic cavernous hemangioma (HCH) and hepatocellular carcinoma (HCC). Recently, convolutional neural networks (CNN) have been widely used in the study of medical image classification because more discriminative image features can be extracted than the human eye. Therefore, this study focused on developing a CNN model for identifying HCH and HCC.

METHODS

This study is a retrospective study. A dataset consisting of 774 non-contrast CT images was collected from 50 patients with HCC or HCH, and the ground truth was given by three radiologists based on contrast-enhanced CT. Firstly, the non-contrast CT images dataset were randomly divided into a training set (n=559) and a test set (n=215). Then, we performed preprocessing of the non-contrast CT images using pseudo-color conversion, and the proposed CNN model developed using training set. Finally, the following indicators (accuracy, precision, recall) were used to quantitatively analyze the results.

RESULTS

In the test set, the proposed CNN model achieved a high classification accuracy of 84.25%, precision of 81.36%, and recall of 82.18%.

CONCLUSIONS

The CNN model for identifying HCH and HCC improves the accuracy of diagnosis on non-contrast CT images.

Isolated complete caudate lobectomy with Glissonean pedicle isolation using Takasaki's technique and right-left approach: preliminary experience from two case reports

Background

Tumors located in the caudate lobe may be primary tumor or metastases from other sites. Isolated caudate lobectomy (ICL) is a challenging procedure due to its complex structure and location. The access route to the caudate lobe has an important role in the success of the operation.

Methods

Based on the characteristics of the segment I location, which is the part of the liver located in front of the vena cava, below the hepatic veins, and cranial to the hilar plate, our approach aims to isolate the entire caudate lobe from these anatomical structures with the following steps: dissecting the caudate lobe from the hilar plate and isolating the caudate lobe from the IVC and from the hepatic veins along with parenchymal resection.

Results

We report two successful cases with the Glissonean pedicle transection method described by Takasaki and the combined right- and left-side approach: a 63-year-old female patient with a 46-mm-in-diameter HCC tumor and a 39-year-old female patient with a 45-mm lesion and the pathological result was focal nodular hyperplasia.

Conclusions

We found this to be a safe and effective approach, which can be applied to all cases of benign tumors or in the case of malignant tumors located entirely in the caudate lobe when extended hepatic resection is not possible due to poor liver function or small remnant liver volume.

First Report of a Late Lethal Pulmonary Biliary Embolism following Hepatic Microwave Ablation

Microwave ablation (MWA) is an established tool in modern therapy of hepatic malignomas. Although it is generally a safe procedure, severe complications related to MWA have been reported in the literature. We report on the first case of a fatal pulmonary biliary embolism following hepatic MWA. The development of pulmonary biliary embolisms is possible and should be considered particularly in the case of by extensive ablation near liver veins.

Artificial intelligence (AI) models for the ultrasonographic diagnosis of liver tumors and comparison of diagnostic accuracies between AI and human experts

BACKGROUND

Ultrasonography (US) is widely used for the diagnosis of liver tumors. However, the accuracy of the diagnosis largely depends on the visual perception of humans. Hence, we aimed to construct artificial intelligence (AI) models for the diagnosis of liver tumors in US.

METHODS

We constructed three AI models based on still B-mode images: model-1 using 24,675 images, model-2 using 57,145 images, and model-3 using 70,950 images. A convolutional neural network was used to train the US images. The four-class liver tumor discrimination by AI, namely, cysts, hemangiomas, hepatocellular carcinoma, and metastatic tumors, was examined. The accuracy of the AI diagnosis was evaluated using tenfold cross-validation. The diagnostic performances of the AI models and human experts were also compared using an independent test cohort of video images.

RESULTS

The diagnostic accuracies of model-1, model-2, and model-3 in the four tumor types are 86.8%, 91.0%, and 91.1%, whereas those for malignant tumor are 91.3%, 94.3%, and 94.3%, respectively. In the independent comparison of the AIs and physicians, the percentages of correct diagnoses (accuracies) by the AIs are 80.0%, 81.8%, and 89.1% in model-1, model-2, and model-3, respectively. Meanwhile, the median percentages of correct diagnoses are 67.3% (range 63.6%-69.1%) and 47.3% (45.5%-47.3%) by human experts and non-experts, respectively.

CONCLUSION

The performance of the AI models surpassed that of human experts in the four-class discrimination and benign and malignant discrimination of liver tumors. Thus, the AI models can help prevent human errors in US diagnosis.

Role of endoscopic ultrasound in the field of hepatology: Recent advances and future trends

The role of endoscopic ultrasound (EUS) as a diagnostic and therapeutic modality for the management of various gastrointestinal diseases has been expanding. The imaging or intervention for various liver diseases has primarily been the domain of radiologists. With the advances in EUS, the domain of endosonologists is rapidly expanding in the field of hepatology. The ability to combine endoscopy and sonography in one hybrid device is a unique property of EUS, together with the ability to bring its probe/transducer near the liver, the area of interest. Its excellent spatial resolution and ability to provide real-time images coupled with several enhancement techniques, such as contrast-enhanced (CE) EUS, have facilitated the growth of EUS. The concept of “Endo-hepatology” encompasses the wide range of diagnostic and therapeutic procedures that are now gradually becoming feasible for managing various liver diseases. Diagnostic advancements can enable a wide array of techniques from elastography and liver biopsy for liver parenchymal diseases, to CE-EUS for focal liver lesions to portal pressure measurements for managing various liver conditions. Similarly, therapeutic advancements range from EUS-guided eradication of varices, drainage of bilomas and abscesses to various EUS-guided modalities of liver tumor management. We provide a comprehensive review of all the different diagnostic and therapeutic EUS modalities available for the management of various liver diseases. A synopsis of all the technical details involving each procedure and the available data has been tabulated, and the future trends in this area have been highlighted.

Liver tumors in children with chronic liver diseases

Liver tumors are rare in children, but the incidence may increase in some circumstances and particularly in chronic liver diseases. Most liver tumors consequent to chronic liver diseases are malignant hepatocellular carcinoma. Other liver tumors include hepatoblastoma, focal nodular hyperplasia, adenoma, pseudotumor, and nodular regenerative hyperplasia. Screening of suspected cases is beneficial. Imaging and surrogate markers of alpha-fetoprotein are used initially as noninvasive tools for surveillance. However, liver biopsy for histopathology evaluation might be necessary for patients with inconclusive findings. Once the malignant liver tumor is detected in children with cirrhosis, liver transplantation is currently considered the preferred option and achieves favorable outcomes. Based on the current evidence, this review focuses on liver tumors with underlying chronic liver disease, their epidemiology, pathogenesis, early recognition, and effective management.

Huge solitary necrotic nodule of the liver: a rare case report with review of literature

Solitary necrotic nodule of the liver (SNNL) is an uncommon disease in clinical practice, and its pathogenesis is still unclear. Here, we report the case of a 35-year-old woman. After physical examination, the patient was found to have a liver neoplasm, and there were no other physical complaints. Abdominal contrast-enhanced computed tomography (CT) showed the presence of a hypodense lesion. The patient opted for surgery to eliminate the lesion. Pathologic examination revealed an isolated necrotic nodular lesion with a size of 12 cm×10 cm×10 cm. The patient had a history of hepatitis B infection. To our knowledge, this is the largest SNNL ever reported and the first case with a history of hepatitis B infection.

Applying criteria for hepatocellular neoplasm of uncertain malignant potential reclassifies more than half of hepatocellular adenomas

OBJECTIVES

Hepatocellular adenoma (HCA) is a benign well-differentiated hepatocellular neoplasm that can be difficult to distinguish from well-differentiated hepatocellular carcinoma (HCC). The term "well-differentiated hepatocellular neoplasm of uncertain malignant potential" (HUMP) has been proposed for neoplasms resembling HCAs, but arising in atypical clinical situations (in females over 50 years old or under 15, in males, with anabolic steroid use, or in some congenital conditions), and/or with atypical pathological features (focal cytological/architectural atypia, β-catenin activation, or focal reticulin loss) insufficient for an unequivocal diagnosis of HCC.

METHODS

This study evaluated HUMP criteria on 42 previously diagnosed HCAs from 33 patients.

RESULTS

Twenty-six (62%) masses from 21 patients were classified as HUMPs. Eleven (42%) had focal cytological atypia, and two (8%) had focal architectural atypia. Four (15%) showed focal reticulin loss. Five (19%) showed evidence of β-catenin activation. Four (12%) HUMP patients were male.

CONCLUSIONS

In this series, HUMP did not correlate with an increased rate of synchronous or metachronous HCC compared to HCA. Clinical colleagues may not accept such a high rate of tumors placed in a category of "uncertain malignant potential". Additional study is warranted to refine criteria for designating well-differentiated hepatocellular neoplasms as of uncertain malignant potential.

Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network

BACKGROUND

Gastroenteropancreatic neuroendocrine tumors most commonly metastasize to the liver; however, high normal background ⁶⁸Ga-DOTATATE activity and high image noise make metastatic lesions difficult to detect. The purpose of this study is to develop a rapid, automated and highly specific method to identify ⁶⁸Ga-DOTATATE PET/CT hepatic lesions using a 2D U-Net convolutional neural network.

METHODS

A retrospective study of ⁶⁸Ga-DOTATATE PET/CT patient studies (n = 125; 57 with ⁶⁸Ga-DOTATATE hepatic lesions and 68 without) was evaluated. The dataset was randomly divided into 75 studies for the training set (36 abnormal, 39 normal), 25 for the validation set (11 abnormal, 14 normal) and 25 for the testing set (11 abnormal, 14 normal). Hepatic lesions were physician annotated using a modified PERCIST threshold, and boundary definition by gradient edge detection. The 2D U-Net was trained independently five times for 100,000 iterations using a linear combination of binary cross-entropy and dice losses with a stochastic gradient descent algorithm. Performance metrics included: positive predictive value (PPV), sensitivity, F₁ score and area under the precision-recall curve (PR-AUC). Five different pixel area thresholds were used to filter noisy predictions.

RESULTS

A total of 233 lesions were annotated with each abnormal study containing a mean of 4 ± 2.75 lesions. A pixel filter of 20 produced the highest mean PPV 0.94 ± 0.01. A pixel filter of 5 produced the highest mean sensitivity 0.74 ± 0.02. The highest mean F₁ score 0.79 ± 0.01 was produced with a 20 pixel filter. The highest mean PR-AUC 0.73 ± 0.03 was produced with a 15 pixel filter.

CONCLUSION

Deep neural networks can automatically detect hepatic lesions in ⁶⁸Ga-DOTATATE PET. Ongoing improvements in data annotation methods, increasing sample sizes and training methods are anticipated to further improve detection performance.