Feasibility of ultrasound radiomics based models for classification of liver fibrosis due to Schistosoma japonicum infection.

BACKGROUND: Schistosomiasis japonica represents a significant public health concern in South Asia. There is an urgent need to optimize existing schistosomiasis diagnostic techniques. This study aims to develop models for the different stages of liver fibrosis caused by Schistosoma infection utilizing ultrasound radiomics and machine learning techniques. METHODS: From 2018 to 2022, we retrospectively collected data on 1,531 patients and 5,671 B-mode ultrasound images from the Second People's Hospital of Duchang City, Jiangxi Province, China. The datasets were screened based on inclusion and exclusion criteria suitable for radiomics models. Liver fibrosis due to Schistosoma infection (LFSI) was categorized into four stages: grade 0, grade 1, grade 2, and grade 3. The data were divided into six binary classification problems, such as group 1 (grade 0 vs. grade 1) and group 2 (grade 0 vs. grade 2). Key radiomic features were extracted using Pyradiomics, the Mann-Whitney U test, and the Least Absolute Shrinkage and Selection Operator (LASSO). Machine learning models were constructed using Support Vector Machine (SVM), and the contribution of different features in the model was described by applying Shapley Additive Explanations (SHAP). RESULTS: This study ultimately included 1,388 patients and their corresponding images. A total of 851 radiomics features were extracted for each binary classification problems. Following feature selection, 18 to 76 features were retained from each groups. The area under the receiver operating characteristic curve (AUC) for the validation cohorts was 0.834 (95% CI: 0.779-0.885) for the LFSI grade 0 vs. LFSI grade 1, 0.771 (95% CI: 0.713-0.835) for LFSI grade 1 vs. LFSI grade 2, and 0.830 (95% CI: 0.762-0.885) for LFSI grade 2 vs. LFSI grade 3. CONCLUSION: Machine learning models based on ultrasound radiomics are feasible for classifying different stages of liver fibrosis caused by Schistosoma infection.

The prevalence of liver abnormalities in humans due to Schistosoma japonicum by ultrasonography in China: a meta-analysis.

BACKGROUND: Schistosoma japonicum was once one of the most severe parasitic diseases in China. After 70 years of national schistosomiasis control programmes, the prevalence and associated morbidity of the infection have been reduced to a much lower level. However, due to the low sensitivity of the current detection approaches, many minor infections in humans could not be identified and ultimately develop chronic injuries with liver abnormalities, a specific 'network' echogenic pattern under ultrasonography. Therefore, as more people take part in physical examinations, we performed this meta-analysis to estimate the overall prevalence of schistosomiasis-associated liver abnormalities in China. METHODS: The publications were searched systematically across five electronic databases. All eligible studies were assessed with quality evaluation forms. Heterogeneity of studies was determined using the I2 and Q tests. A random effects or fixed effects model was employed based on heterogeneity results. The pooled prevalence and its 95% confidence intervals were calculated with the Freeman-Tukey double arcsine transformation. All analyses were conducted using R with the "meta" package. The protocol registration number was CRD42021232982. RESULTS: A total of 19 relevant articles, including 21 studies, were included. The average score of study quality was 6.4 (total score 7), indicating high quality of all included studies. A total of 268, 247 persons were included, and 43, 917 persons were diagnosed with schistosomiasis liver abnormalities by ultrasonography. High degrees of heterogeneity existed among all studies or within subgroups. The overall pooled prevalence was 18.64% (95% CI: 11.88-26.50%). The estimate significantly increased over time and varied among provinces, with the highest in Shanghai and the lowest in Sichuan. The estimate in people aged 60 years or older was significantly higher than that in people of all ages. No significant difference was seen when based on study areas (urban or rural areas) or gender. CONCLUSION: The long-term burden of schistosomiasis in China remains large, as nearly one-fifth of the examined persons were diagnosed with schistosomiasis liver abnormalities. The pooled prevalence was associated with regions or age groups. Such may have a high reference value in the exact calculation of the disease burden and can be helpful for policy makers in prioritizing public health.

Analysis of CT characteristics in the diagnosis of Schistosoma japonicum associated appendicitis with clinical and pathological correlation: a diagnostic accuracy study.

PURPOSE: To clarify unique non-contrast CT (NCCT) characteristics for early recognition of Schistosomal associated appendicitis (SAA) differentiating from Non-schistosomal associated appendicitis (NSA). MATERIAL AND METHODS: Clinical and pathological data of 50 cases with SAA and 60 cases with NSA who underwent emergency appendectomy were retrospectively compared to pre-surgical NCCT features such as direct and indirect signs of acute appendicitis as well as appendicoliths, colon calcifications as diagnostic criteria. Statistical methods such as Chi-square (χ2), t-tests, Principal component analysis (PCA), Binary Logistic regression (LR) and Factor Analysis (FA) were utilized to observe differences and isolate recognizable CT features of SAA. Pre and post hoc diagnostic performance of all criteria was calculated as sensitivity, specificity, and the Odds Ratio (OR). RESULTS: Age > 50 years, diameter > 13 mm, pneumatosis, peri appendiceal abscess, focal wall defect, perforation; Orbital, linear and point types of appendicular wall calcifications; sigmoid colon and cecal curvilinear calcifications were observed as unique characteristics with a sensitivity of 84-95% and specificity of 91-98% in predicting SAA by OR of 6.2 times. Pre and post hoc hypothetical analysis did not show any significance for all other factors. CONCLUSION: Factors such as elderly age, CT features such as larger appendicular diameter, appendicular wall calcifications along with sigmoid colon, and cecal calcifications, signs of perforation or abscess are characteristic for early recognition of SAA.

Lessons from a 15-year-old boy with advanced schistosomiasis japonica in China: a case report.

Schistosomiasis is a chronic, parasitic disease caused by flukes (trematodes) of the genus Schistosoma, which presents the most important global burden of the 17 neglected tropical diseases listed by the World Health Organization. China has made great achievements in schistosomiasis control, and now China is planning to move forward, to eliminate schistosomiasis within 2020, but the fact cannot be denied that the possibility of schistosome infection is still there in some endemic due to its zoonotic nature as well as wide distribution of its intermediate hosts (snails). Thus, how to interrupt the transmission in areas with distribution of schistosomes and intermediate snails becomes a very serious challenge that China is facing. In this paper, it is reported an advanced schistosomiasis japonica case of a 15-year-old boy which is extremely rare in the current schistosomiasis control in China. Thus, it is supposed to strengthen health education of school children and to train professional physicians of local hospitals.

[MRI findings of cerebral schistosomiasis in acute stage：establishment of experimental model of acute cerebral schistosomiasis with rabbits].

OBJECTIVE: To establish an experimental model of acute cerebral schistosomiasis japonica and explore the MRI manifestations of acute cerebral schistosomiasis. METHODS: Rabbits were divided into 3 groups with 10 rabbits in each group. The rabbits in the experimental group were directly injected with suspension fluid of Schistosoma japonicum eggs (0.9 mg, 1 ml) by the cranial drilling method, those in the negative control group were given saline (1 ml) by the same method above-mentioned, and those in the blank control group were not given any treatment. Antibiotic was given to the first two groups after the operation. The clinical manifestations of the 3 groups were observed, and the magnetic resonance imaging (MRI) was performed in 30 days post-operation, and then the brain tissues were taken for pathological examinations. RESULTS: All the rabbits in the experimental group exhibited inappetence, various neurological symptoms including hemiplegia, and weight loss after the operation; while those in the negative control group showed inappetence in 3 days after the operation, and 1 week later, the symptom disappeared; there were no adverse reactions in the blank control group. MRI of the experimental group showed nodular or patchy enhancement on T1WI enhancement, brain edema, abnormal ventricular dilatation, and needle augmentation. SWI displayed hypointense in the abnormal enhanced nodules and flaky hypointense on the operation brain. In the negative control group, 2 rabbits showed abnormal enhancement of the needle canal, and 1 showed mild dilatation of the ventricle. The blank control group showed normal manifestations. The pathological examinations showed abnormal appearances in 10 rabbits of the experimental group, including 6 with S. japonicum egg granuloma nodules, nonspecific granuloma nodules coexisted with perivascular inflammation; no granuloma nodules were found in the negative control group, but 2 rabbits showed vascular inflammation; the blank control group showed the normal brain tissue. CONCLUSIONS: An experimental model of acute cerebral schistosomiasis is successfully established in rabbits by intracranial injection of schistosome eggs. The MRI examination combined with the clinical manifestations can improve the accuracy of early diagnosis of cerebral schistosomiasis.

Magnetic Resonance Spectroscopy for Evaluating Portal-Systemic Encephalopathy in Patients with Chronic Hepatic Schistosomiasis Japonicum.

Portal-systemic encephalopathy (PSE) is classified as type B hepatic encephalopathy. Portal-systemic shunting rather than liver dysfunction is the main cause of PSE in chronic hepatic schistosomiasis japonicum (HSJ) patients. Owing to lack of detectable evidence of intrinsic liver disease, chronic HSJ patients with PSE are frequently clinically undetected or misdiagnosed, especially chronic HSJ patients with covert PSE (subclinical encephalopathy). In this study, we investigated whether magnetic resonance spectroscopy (MRS) could be a useful tool for diagnosing PSE in chronic HSJ patients. Magnetic resonance (MR) T1-weighted imaging, diffusion-weighted imaging, and MRS were performed in 41 chronic HSJ patients with suspected PSE and in 21 age-matched controls. The T1 signal intensity index (T1SI) and apparent diffusion coefficient (ADC) value were obtained in the Globus pallidus. Liver function was also investigated via serum ammonia and liver function tests. Higher T1SI and ADC values, increased lactate and glutamine levels, and decreased myo-inositol were found in the bilateral Globus pallidus in chronic HSJ patients with PSE. No significantly abnormal serum ammonia or liver function tests were observed in chronic HSJ patients with PSE. On the basis of these findings, we propose a diagnostic procedure for PSE in chronic HSJ patients. This study reveals that MRS can be useful for diagnosing PSE in chronic HSJ patients.

Tissue inhibitor of matrix-metalloprotease-1 predicts risk of hepatic fibrosis in human Schistosoma japonicum infection.

BACKGROUND: Schistosomes infect 200 million individuals annually and cause significant hepatic fibrosis in up to 20%. Little is known regarding the mechanisms of schistosome-associated hepatic fibrosis in humans, and few biomarkers for risk of fibrosis have been identified. METHODS: We treated 611 Schistosoma japonicum-infected Filipinos with praziquantel (PZQ) and performed ultrasound to quantify hepatic fibrosis at baseline and 12 months after PZQ treatment. We developed a multiplexed assay (FibroPlex) that quantifies predictors and effect modifiers of fibrosis. We measured FibroPlex analytes produced by peripheral blood mononuclear cells stimulated with schistosome egg antigen 4 weeks after PZQ treatment and related these levels to risk of fibrosis 1 year after PZQ treatment. RESULTS: After adjusting for potential confounders, including baseline grade of fibrosis, individuals with detectable tissue inhibitor of matrix-metalloprotease-1 (TIMP-1) had a 3.5-fold greater risk of fibrosis 1 year after PZQ treatment, compared with individuals with undetectable levels (odds ratio, 3.48; 95% confidence interval, 1.41-8.43; P = .007). DISCUSSION: Because TIMP-1 inhibits most matrix metalloproteases, which are responsible for collagen degradation, these data suggest that schistosome-associated hepatic fibrosis results, in part, from excessive inhibition of collagen remodeling. These data further suggest that TIMP-1 is a promising biomarker for assessing risk of hepatic fibrosis in schistosomiasis and, potentially, other infectious and noninfectious causes of liver disease.

Severe hepatosplenic schistosomiasis: clinicopathologic study of 102 cases undergoing splenectomy.

We present the preoperative findings of 102 patients who underwent successful splenectomy for advanced schistosomiasis japonica. All patients were symptomatic for schistosomiasis and had splenomegaly greater than or equal to II according to the Hackett criteria. Before surgery, all patients underwent clinical examination including full blood count; fibrinogen and serum protein levels; liver function tests; and serology for hepatitis B, C, and D. Ultrasound examination of the liver and spleen and liver histology for evidence of pathology were also undertaken. Ninety patients had a treatment history for schistosomiasis. Fifty-six patients were seropositive for hepatitis B virus antibody, and 6 patients were seropositive for hepatitis C virus antibody. Immunohistochemical testing of the liver samples confirmed that 45 patients were positive for hepatitis B virus surface antigen, thereby indicating active infection. A total of 66.7% of patients had fibrosis stages II to III by ultrasound; and 76.5% of patients had portal vein inner diameter greater than 12 mm, indicating portal vein hypertension. A total of 83.2% of patients showed various stages of esophageal varicosis via x-ray, and 81.4% had fibrotic stages III to IV by liver biopsy. Coinfection with hepatitis B virus accelerated the development of liver fibrosis. There was moderate concordance between the fibrosis assessed by ultrasonography and histopathology, indicating that ultrasound underestimates the true pathology. Combined assessment is needed to improve the diagnosis of clinical hepatic fibrosis.

Cerebellar schistosomiasis: a case report with clinical analysis.

The authors report here a rare case of cerebellar schistosomiasis identified by pathological diagnosis, lacking extracranial involvement. The clinical symptoms included headache, dizziness, and nausea. Studies in blood were normal and no parasite eggs were detected in stool. Computed tomography of brains showed hypodense signal, and magnetic resonance imaging showed isointense signal on T1-weighted images, hyperintense signal on T2-weighted images, and intensely enhancing nodules in the right cerebellum after intravenous administration of gadolinium. A high-grade glioma was suspected, and an operation was performed. The pathologic examination of the biopsy specimen revealed schistosomal granulomas scattered within the parenchyma of the cerebellum. The definitive diagnosis was cerebellar schistosomiasis japonica. A standard use of praziquantel and corticosteroid drugs was applied, and the prognosis was good. When the pattern of imaging examinations is present as mentioned above, a diagnosis of brain schistosomiasis should be considered.

Radiology corner. Answer to last month's radiology case and images: Schistosomiasis japonicum involving the liver and colon.

Abdominal radiography showed subtle curvilinear densities overlying the left lower quadrant and central within the pelvis (Fig. 1a). This was confirmed on intravenous contrast-enhanced abdominopelvic CT to be caused by thin mucosal surface calcifications within the descending and sigmoid colons (Fig. 1b, c). Incidental note is made of thin right hepatic lobe capsule calcifications (Fig. 1d). These colonic and hepatic calcifications are characteristic of Schistosomiasis infection of the gastrointestinal tract. Liver biopsy confirmed Schistosomiasis japonicum infection.

[Validity of inquiry in screening chronic schistosomiasis japonica].

OBJECTIVE: To study the indicators and validity of inquiry in the screening of chronic schistosomiasis japonica. METHODS: 51 villages of Hanshou county were selected at random in Hunan Province and the whole resident (>6 months/year) population aged 5 years and above was eligible for inclusion in the study. Inquiry through questionnairing, serological test (ELISA) and B type ultrasonography were applied respectively to screen chronic cases and evaluate morbidity due to schistosome infection. Logistic analysis was performed to explore the relationship between indicators of questionnaire and the results of ELISA and abdominal ultrasonography. Bayes discriminant analysis was used to assess consistency of inquiry and ELISA, inquiry and the degree of hepatic fibrosis. RESULTS: 26,426 inhabitants in the endemic villages were screened by ELISA with 1380 (5.2%) positive. 1264 sero-positive and 1446 sero-negative cases were asked questions relating to schistosomiasis and examined by abdominal ultrasonography. Inquiry indices such as self-reported diarrhea, stool with mucus and fatigue during the last two weeks, history of infested water contact and times of treatment were specific to chronic schistosomiasis. The coincident rate for validation was 75.9% between inquiry and ELISA, and 75.4% between inquiry and hepatic fibrosis degree. CONCLUSION: Validity of inquiry was satisfactory in screening chronic schistosomiasis in endemic areas.

Determination of the period for establishment of a liver network echogenic pattern in Schistosoma japonicum infection.

Schistosomiasis is caused by infection with Schistosoma haematobium, S. mansoni, S. japonicum, or S. mekongi. S. japonicum infection results in liver cirrhosis at the final stage. A "network" (NW) echogenic pattern on hepatic ultrasonography appears to be specific to S. japonicum infection. The principal aim of the present study was to determine the exact year(s) or even month(s) required for the establishment of the liver NW echogenic pattern from the initial infection in young patients with schistosomiasis japonica since there are few data on this important point. We conducted yearly ultrasonographic, serologic, coprologic, and physical examinations of schistosomiasis patients in the Philippines from 1996 up to the present. During that period, the total number of patients examined was approximately 2,000, among whom we selected 2 patients for determination of the duration required for NW establishment, when they were 10 years old. Although the exact time of initial exposure to schistosomes cannot be determined, the duration for the establishment of NW was definitively confirmed in patient no. 1 to be between 19-24 months based on the results of serologic and coprologic examinations. For patient no. 2, the circumstantial evidence suggested that the establishment of a NW might require 5 to 6 years at maximum. To the best of our knowledge, this is the first evidence-based report on the determination of the period required for the establishment of a liver NW echogenic pattern in S. japonicum infection in the Philippines.

Prevalence of schistomasiasis lesions detected by ultrasonography in children in Molodo, Mali.

AIM: To study schistomasiasis infection in school children in Molodo, an irrigated rice growing region of Mali, by determining the prevalence of schistomasiasis and lesions identified by ultrasonography among children living in this region. METHODS: This cross sectional study included 346 children aged 7 to 14 years selected at random from five schools in Molodo. We tested for hematuria using urine dipsticks and searched for Schistosoma haematobium eggs in urine and S. mansoni eggs in stools. Ultrasonography of the liver, spleen and urinary tract was performed. RESULTS: The prevalences of Schistosoma haematobium and S. mansoni infection were 72% (range: 66.9-76.6%) and 68.2% (range: 60.9-71.2%) respectively; 55.1% of the children had co-infection. Ultrasonography of the urinary tract revealed an irregular bladder wall as the most frequent abnormality (3.4% of children). Abdominal ultrasonography demonstrated type B hepatic fibrosis in four children (1.1%), type C in one (0.3%) and type D in one (0.3%). CONCLUSION: Few schistosomiasis lesions were detected by ultrasonography compared with the prevalence of S. haematobium and S. mansoni infections. This observation is probably related to mass treatment programs conducted during a national anti-schistosomiasis program.

[Canonical correlation and redundancy analysis on the indices of abdominal ultrasonography of schistosomiasis japonica].

OBJECTIVE: To explore the correlation of the ultrasound indices of liver and spleen in schistosomiasis japonica and with infection frequency, infection time and EPG. METHODS: The canonical correlation was applied to analyze the correlation of the hepatosplenic ultrasound indices in schistosomiasis japonica, and the correlation of the ultrasound indices with infection frequency, infection time and EPG. The proportions of variation related to each other in liver and spleen were analyzed by the redundancy analysis. RESULTS: The correlation coefficients of the first pair correlation canonical variable were 0.7842, 0.5483 and 0.5800, 0.4220, respectively, in males and females without infection, males and females with infection history (P<0.01). The correlation coefficients of the first pair correlation canonical variable were 0.6063, 0.5215 and 0.6595, 0.3849, respectively, in male negatives and female negatives, male positives and female positives (P<0.01). In groups of males and females without infection, the variations of liver ultrasound indices related with the variation of spleen ultrasound indices were 43.5% and 17.5% respectively, and in groups infection history, they were 22.1% and 11.4% respectively. In male and female negative groups, the variations of liver ultrasound indices related with the variation of spleen ultrasound indices were 26.8% and 16.8% respectively, and in positive groups, they were 27.6% and 10.7% respectively. The infection frequency, infection time and EPG in the stool-positive groups were not significantly related with the canonical variables of liver and spleen ultrasound indices (P>0.05). CONCLUSION: There is a significant canonical correlation between the liver and spleen ultrasound indices in schistosomiasis japonica, especially in males. In groups with infection history, the infection frequency, infection time and EPG of those stool-positives are not relevant to the canonical variables of liver and spleen ultrasound indices.

Imaging diagnosis of schistosomiasis japonica--the use in Japan and application for field study in the present endemic area.

For detecting lesions-related schistosomiasis japonica, X-rays, scintillation scanning, ultrasonography (US), computed tomography (CT), magnetic resonance (MR) and endoscopic examinations with biopsies have been used in Japan. Liver fibrosis and calcified changes are detected by US and CT. Most of the lesions that are detected by endoscopic examinations are due to deposited ova of Schistosoma japonicum. Portal hypertension is detected by US, CT and gastroscopic examination. Because schistosome infection decreased rapidly in Japan, most of the studies on imaging diagnosis were performed on chronic lesions or sequelae of schistosomiasis. Most of the techniques were used on admitted patients in well-equipped hospitals. US was introduced in the 1970s as a safe, rapid, non-invasive and inexpensive technique and has been used for diagnosis in hospitals and screening in the fields. As a typical US image of the liver, septal formation by high echogenic bands like mosaic was described, and this network pattern was reported in the other endemic countries; China and Philippines. As an appropriate technique, US has been broadly used in developing countries. Not only for diagnosis in a hospital, but also for monitoring changes of morbidity, US is used in the community level. Network pattern related to the severity of S. japonicum infection, has not been described in S. mansoni or S. haematobium infection. Appearance of network pattern depends on pathological changes such as periportal fibrosis, postnecrotic fibrosis and calcified ova. For advanced studies on morbidity of schistosomiasis japonica, further research on pathological basis of network pattern and standardization of US diagnosis are necessary.

[Principal component analysis on ultrasound indexes of schistosomiasis and the assessment on prevalence rate].

OBJECTIVE: To explore the synthetical index for diagnosing schistosomiasis with ultrasound and to assess the prevalence rate with the index. METHODS: Ultrasound indexes of schistosomiasis Japonicum were analyzed by principal component analysis, and the synthetical indexes were assessed by ROC curve. RESULTS: Among the abnormal rates of the 6 indexes, the lowest was 1.6% comparing with the highest of 59.5%. Significant difference was noficed among the abnormal rates (chi(2) = 631.1, P < 0.01). The individual correlation of the six indexes to each other as will as with age distribution was significant (P < 0.05). The three principal components reflected the degree of pathological changes on liver and spleen. The first principal component was the factor reflecting the degree of liver pathological changes, and the second and third principal components reflected the degree of pathological changes on spleen. The synthetical index D(1) = 0.047X(1) + 0.428X(2) + 1.247X(3) + 0.095X(4) + 0.002X(5) + 0.213X(6) - 12.837 was found by adding the three weight principal components, and it's area under the ROC curve was 0.957. When -1.70 was taken as the critical value, the abnormal rate of population was 66.3%, close to the resident's actual prevalence rate 66.9%. CONCLUSION: Ultrasonography was considered as a method which could rapidly assessing the resident's prevalence rate in the endemic areas of schisitosomiasis Japonicum, and could also provide powerful information for development of strategy on chemotherapy.