Ultrasound identification of hepatic echinococcosis using a deep convolutional neural network model in China: a retrospective, large-scale, multicentre, diagnostic accuracy study

BACKGROUND

Ultrasonography is the most widely used technique to diagnose echinococcosis; however, challenges in using this technique and the demand on medical resources, especially in low-income or remote areas, can delay diagnosis. We aimed to develop a deep convolutional neural network (DCNN) model based on ultrasonography to identify echinococcosis and its types, especially alveolar echinococcosis.

METHODS

This retrospective, large-scale, multicentre study used ultrasound images from patients assessed at 84 hospitals in China, obtained between Jan 1, 2002, and Dec 31, 2021. Patients with a diagnosis of cystic echinococcosis, alveolar echinococcosis, or seven other types of focal liver lesions were included. We tested ResNet-50, ResNext-50, and VGG-16 as the backbone network architecture for a classification DCNN model and input the perinodular information from the ultrasound images. We trained and validated the DCNN model to diagnose and classify echinococcosis using still greyscale ultrasound images of focal liver lesions in four stages: differentiating between echinococcosis and other focal liver lesions (stage one); differentiating cystic echinococcosis, alveolar echinococcosis, and other focal liver lesions (stage two); differentiating cystic echinococcosis, alveolar echinococcosis, benign other focal liver lesions, and malignant focal liver lesions (stage three); and differentiating between active and transitional cystic echinococcosis and inactive cystic echinococcosis (stage four). We then tested the algorithm on internal, external, and prospective test datasets. The performance of DCNN was also compared with that of 12 radiologists recruited between Jan 15, 2022, and Jan 28, 2022, from Qinghai, Xinjiang, Anhui, Henan, Xizang, and Beijing, China, with different levels of diagnostic experience for echinococcosis and other focal liver lesions in a subset of ultrasound data that were randomly chosen from the prospective test dataset. The study is registered at ClinicalTrials.gov (NCT03871140).

FINDINGS

The study took place between Jan 1, 2002, and Dec 31, 2021. In total, to train and test the DCNN model, we used 9631 liver ultrasound images from 6784 patients (2819 [41·7%] female patients and 3943 [58·3%] male patients) from 87 Chinese hospitals. The DCNN model was trained with 6328 images, internally validated with 984 images, and tested with 2319 images. The ResNet-50 network architecture outperformed VGG-16 and ResNext-50 and was generalisable, with areas under the receiver operating characteristic curve (AUCs) of 0·982 (95% CI 0·960-0·994), 0·984 (0·972-0·992), and 0·913 (0·886-0·935) in distinguishing echinococcosis from other focal liver lesions; 0·986 (0·966-0·996), 0·962 (0·946-0·975), and 0·900 (0·872-0·924) in distinguishing alveolar echinococcosis from cystic echinococcosis and other focal liver lesions; and 0·974 (0·818-1·000), 0·956 (0·875-0·991), and 0·944 (0·844-0·988) in distinguishing active and transitional cystic echinococcosis from inactive echinococcosis in the three test datasets. Specifically, in patients with the hepatitis B or hepatitis C virus, the model could distinguish alveolar echinococcosis from hepatocellular carcinoma with an AUC of 0·892 (0·812-0·946). In identifying echinococcosis, the model showed significantly better performance compared with senior radiologists from a high-endemicity area (AUC 0·942 [0·904-0·967] vs 0·844 [0·820-0·866]; p=0·027) and improved the diagnostic ability of junior, attending, and senior radiologists before and after assistance with AI with comparison of AUCs of 0·743 (0·714-0·770) versus 0·850 (0·826-0·871); p<0·0001, 0·808 (0·782-0·832) versus 0·886 (0·864-0·905); p<0·0001, and 0·844 (0·820-0·866) versus 0·870 (0·847-0·890); p=0·092, respectively.

INTERPRETATION

The DCNN model was shown to be accurate and robust, and could improve the ultrasound diagnostic ability of radiologists for echinococcosis and its types for highly endemic and remote regions.

FUNDING

National Natural Science Foundation of China and National Key Research & Development Program of China.

TRANSLATION

For the Chinese translation of the abstract see Supplementary Materials section.

Analysis of the efficacy of microwave ablation in the treatment of early hepatic alveolar echinococcosis: A propensity score matching based study

BACKGROUND

To explore the efficacy of microwave ablation (MWA) in the treatment of hepatic alveolar echinococcosis (HAE) with a diameter of ≤5 cm.

METHOD

From June 2014 to January 2020, patients diagnosed with HAE were retrospectively analyzed. After balancing the confounding factors by propensity score matching (PSM) , the patients were divided into MWA group (n = 20) and radical operation group (n = 20) by 1:1 matching. The safety and effectiveness of MWA were assessed by comparing the differences between the two groups in terms of postoperative general laboratory indices, grading of postoperative complications, length of postoperative hospitalization, the outcome of treatment, and disease recurrence.

RESULT

After PSM, all confounders were not statistically different (P>0.05) . Compared with the radical surgery group, patients in the MWA group had lower postoperative ALT and WBC elevations (P<0.001) , shorter postoperative hospital stay (P<0.001) ) , lower hospital costs (P<0.001) . The effective rate of the two groups was 100%. There was no statistical difference in the incidence of postoperative complications and recurrence rate (P>0.05).

CONCLUSION

MWA is a safe and effective means of treating HAE ≤ 5 cm in diameter.

Effectiveness and safety of ultrasound-guided percutaneous microwave ablation for hepatic alveolar echinococcosis

Background

Microwave ablation (MWA) is a popular therapy for liver malignant tumor in recent years. Few studies have been conducted on its use in the treatment of hepatic alveolar echinococcosis (HAE). The study aims to evaluate the efficacy and safety of MWA in the treatment of HAE.

Methods

This study analyzed the data of 45 patients (mean age, 38 ± 2 years; 24 males) diagnosed with HAE and underwent MWA treatment between June 2014 to December 2019. The patients after MWA were examined by CT or MRI [follow-up: 32 months (IQR 23–48.5)] to determine whether the lesions were relapsed and to evaluate the therapeutic effect of MWA. The safety of MWA was evaluated by monitoring postoperative complications. Clinical data, such as patient demographics, imaging features of the lesions, relevant findings of laboratory tests before and after ablation, and information related to ablation, were collected and analyzed. Paired-sample t tests and paired-sample Wilcoxon signed-rank tests were used to compare relevant laboratory indicators before and after MWA.

Results

MWA was applied to 57 HAE lesions in 45 patients. The median size of lesions was 3.42 cm (IQR2.85–4.41). The rate of complete ablation was 100% (57/57). The median follow-up time was 32 months (IQR 23–48.5). The recurrence rate was 13% (6/45), and the median time of recurrence was 22 months. The rate of minor complications was 11.1% (5/45), and there were no major complications and deaths. Compared to preoperative, ALB, RBC, HBG, and PLT were decreased (p < 0.001); ALT, TB, DB, and WBC were increased (p < 0.001); and no statistically difference in PT, APTT, and INR (p > 0.05).

Conclusions

MWA might be a safe and effective way to cure HAE. Meanwhile, it provides a new option and a new way of thinking about treatment for patients with HAE.

Parasite Circulating Cell-free DNA in the Blood of Alveolar Echinococcosis Patients as a Diagnostic and Treatment-Status Indicator

BACKGROUND

Alveolar echinococcosis (AE) is a serious parasitic disease caused by the larvae of Echinococcus multilocularis. It is the less common but substantially more deadly of the 2 major echinococcosis diseases that can occur globally but are concentrated in central Asia.

METHODS

We analyzed parasite circulating cell-free DNA (cfDNA) in 149 plasma samples using a DNA sequencing-based method (105 AE, 16 cystic echinococcosis, 4 liver cancer, 4 gallstones, and 20 healthy volunteers). After identifying the Echinococcus-specific cfDNA (Em-cfDNA) sequences in the samples, we determined whether Em-cfDNA could be used for AE diagnosis and as a potential indicator of the effectiveness of surgical treatment. We also examined potential associations between Em-cfDNA levels and clinical features of AE patients.

RESULTS

Our work demonstrates that varying reads of Em-cfDNA were detectable in the plasma of 100% of preoperative AE patients and that all of the non-AE patients and healthy volunteers were negative. Em-cfDNA has good sensitivity and specificity for the diagnosis of AE. We also found that Em-cfDNA levels apparently have reference value for evaluating the therapeutic efficacy of surgery interventions for AE lesions. Finally, our analysis revealed that Em-cfDNA levels can reflect meaningful information about lesion size in preoperative AE patients.

CONCLUSIONS

We demonstrate that sequencing-based monitoring of Em-cfDNA can be used in the clinic as a powerful diagnostic indicator for AE. We also note that there is a strong potential for use of this liquid-biopsy method to monitor ongoing disease status in postintervention AE patients.

Efficacy and safety of ultrasound-guided percutaneous microwave ablation for the treatment of hepatic alveolar echinococcosis: A preliminary study

The present study aims to assess the efficacy and safety of ultrasound-guided percutaneous microwave ablation (MWA) for hepatic alveolar echinococcosis (HAE) preliminarily.Seventeen patients diagnosed to HAE and treated with MWA (80 watts, 4 min) were retrospectively analyzed. The upper abdominal computed tomography (CT) was performed at 1, 6, 12 months after the MWA treatment. The complications were evaluated to assess the safety.The diameters of the lesions in the HAE patients ranged from 1.9 to 4.7 cm. The patients included 10 males and 7 females, aged 26 to 70 (45.82 ± 13.36) years, 5 patients infecting with chronic hepatitis viral B and 8 patients with positive hydatid antibody (IgG). The lesions observed in the postoperative CT (1, 6, 12 months) were calcified compared with those observed in the preoperative CT and without relapse. No serious treatment-related complications occurred after treatment.MWA is a novel and effective therapeutic method for HAE with a single lesion (diameter≤=5 cm). Further studies based on prospective random control trials to confirm our findings are necessary.