USING A FUZZY ENGINE AND COMPLETE SET OF FEATURES FOR HEPATIC DISEASES DIAGNOSIS: INTEGRATING CONTRAST AND NON-CONTRAST CT IMAGES

In the diagnosis of hepatic diseases, “Contrast-Enhanced Computerized Tomography” (CECT) and “Non-Contrast CT” (NCT) are usually simultaneously adopted. In this paper, a system consisting of a fuzzy diagnosis engine and a feature quantizer, which extracts hepatic features from CECT and NCT images, is proposed for assisting hepatic disease diagnosis. Compared with existing methods, this paper differs in two folds. First, a more complete feature set composed of not only lesion textures, but also lesion morphological structure and lesion contrast to normal tissues is used. These features are described through mathematical models built inside the feature quantizer and served as the input of fuzzy diagnosis engine. Second, because of the use of the fuzzy diagnosis engine, the system is intrinsically with the capability of storing rules and may infer and adapt its rules according to learning data. Furthermore, uncertainty associated with disease diagnosis can be appropriately taken into considerations. The system has been tested using 131 sets of image data, which are to be classified into 4 types of diseases: liver cyst, hepatoma, cavernous hemagioma and metastatic liver tumor. Experimental results indicate that among these test data 78% of them are accurately classified as one type, while the remaining 22% of data are classified as more than one types of diseases. Even so, within these 22% of multiple-classified data, the correct type is always included in the output in each test, showing a promise of the system.

Hemangioma-like lesions in chronic liver disease: diagnostic evaluation in patients

PURPOSE

To quantify the risk of misdiagnosis of focal hepatic lesions manifesting at ultrasonography (US) as typical hemangiomas in a population at high risk for hepatocellular carcinoma (HCC) and to identify the most effective approach to their diagnostic evaluation.

MATERIALS AND METHODS

A total of 1,982 patients with newly diagnosed cirrhosis underwent US and serum alpha-fetoprotein determinations for early detection of HCC. Focal lesions with typical features of hemangioma were evaluated with confirmatory findings of contrast material-enhanced dynamic or spiral computed tomography (CT) and/or single photon emission CT with technetium 99m-labeled red blood cells and, in the absence of confirmatory imaging findings, US-guided fine-needle biopsy. Patients whose initial US scan depicted no lesions or hemangiomas were enrolled in a US follow-up program. All hemangioma-like lesions detected during follow-up were evaluated, or biopsy was performed.

RESULTS

US depicted hemangioma-like lesions in 44 of 1,982 patients: 22 hemangiomas and 22 HCCs. Hemangioma-like lesions detected during follow-up in 1,648 patients were HCCs (n = 22) or dysplastic nodules (n = 4). Only 85 (22%) of 383 patients with HCC had alpha-fetoprotein levels suggestive of the diagnosis. The probability of a diagnosis of HCC (or preneoplastic lesion) is 100% for hemangioma-like lesions depicted on subsequent US scans.

CONCLUSION

If initial US examination of a cirrhotic liver depicts a hemangioma, confirmatory findings of imaging studies are necessary since 50% of hemangiomas in this study were hyperechogenic HCCs. US-guided biopsy can be safely performed, and its findings can be used to confirm the diagnosis.

Hepatic cavernous hemangioma in cirrhotic liver: imaging findings

OBJECTIVE

To document the imaging findings of hepatic cavernous hemangioma detected in cirrhotic liver.

MATERIALS AND METHODS

The imaging findings of 14 hepatic cavernous hemangiomas in ten patients with liver cirrhosis were retrospectively analyzed. A diagnosis of hepatic cavernous hemangioma was based on the findings of two or more of the following imaging studies: MR, including contrast-enhanced dynamic imaging (n = 10), dynamic CT (n = 4), hepatic arteriography (n = 9), and US (n = 10).

RESULTS

The mean size of the 14 hepatic hemangiomas was 0.9 (range, 0.5-1.5) cm in the longest dimension. In 11 of these (79%), contrast-enhanced dynamic CT and MR imaging showed rapid contrast enhancement of the entire lesion during the early phase, and hepatic arteriography revealed globular enhancement and rapid filling-in. On contrast-enhanced MR images, three lesions (21%) showed partial enhancement until the 5-min delayed phases. US indicated that while three slowly enhancing lesions were homogeneously hyperechoic, 9 (82%) of 11 showing rapid enhancement were not delineated.

CONCLUSION

The majority of hepatic cavernous hemangiomas detected in cirrhotic liver are small in size, and in many, hepatic arteriography and/or contrast-enhanced dynamic CT and MR imaging demonstrates rapid enhancement. US, however, fails to distinguish a lesion of this kind from its cirrhotic background.

Iodized-oil retention within hepatic hemangioma: characteristics on iodized-oil CT

OBJECTIVE

The purpose of this study was to describe the characteristic computed tomographic (CT) appearance of iodized-oil retention in hepatic hemangioma and to evaluate the duration of the retention of iodized oil on follow-up CT.

METHODS

Seventeen hepatic hemangiomas of 14 patients were studied with CT performed 1-3 weeks after injection of 2-9 ml of iodized oil (iodized-oil CT) for the characterization of focal hepatic lesions, which needed differential diagnosis with hepatocellular carcinoma in 10 patients, for therapy in two patients, and for chemoembolization therapy of accompanying hepatocellular carcinomas in two. Twelve patients had 1-7 follow-up CT scans within an interval of 1-38 months.

RESULTS

In all cases, iodized-oil CT showed iodized-oil retention within the tumor, regardless of tumor size, shape, location, and amount of injected iodized oil. The distribution was incomplete and predominantly peripheral in all cases. Central retention was also seen in seven cases, in which a relatively large amount of iodized oil was injected, but retention of iodized oil in the tumor was incomplete even in two cases in which a large amount of iodized oil was injected to relieve symptoms and in three cases in which prominent uptake of surrounding liver parenchyma was seen. Patterns of retention were predominantly spotty in five, predominantly nodular in four, and mixed in eight patients. Retention materials slowly washed out but persisted for at least 3 months and up to 38 months (mean = 18.1 months), and complete washout was not seen in any cases at follow-up CT.

CONCLUSION

In all cases of hepatic hemangiomas, iodized oil was retained, and retention persisted over several months. Distribution and patterns of retention were characteristically peripheral, spotty, and nodular at iodized-oil CT. Knowledge of the iodized-oil CT appearance of hepatic hemangioma would be helpful to interpret follow-up CT studies of patients who have undergone iodized-oil chemoembolization procedures.

Hepatocellular carcinoma: prospective assessments of the T-factor with CT, US, and MR imaging

The capabilities of computed tomography (CT), ultrasonography (US), and magnetic resonance (MR) imaging were studied in order to determine the role of each of these noninvasive examinations for estimating the T-factor of hepatocellular carcinomas (HCCs). Fifty-one patients with surgically proven HCCs received CT (50 patients), US (46 patients), and MR (44 patients). The images of CT, US, and MR were prospectively evaluated for main tumor size, intrahepatic metastases, and vascular invasion, which compose the T-factor of HCC, and compared to pathological results. The sizes of the main tumor were estimated correctly by all examinations. For estimating intrahepatic metastases, US (74%) and MR (73%) were superior to CT (65%). For estimating portal invasion, CT (79%) was superior to US (70%) and MR (66%), because CT could demonstrate the segmental staining caused by portal invasion. The estimates of hepatic venous invasion were difficult during any of the examinations. We conclude that presurgical evaluations of the T-factor require the use of US and CT or MR and CT.

Management of haemangioma of the liver: comparison of results between surgery and observation

Cavernous haemangioma of the liver was diagnosed in 33 patients using hepatic angiography between 1978 and 1988. Thirteen of these patients underwent surgery. There were no deaths after operation; morbidity included upper gastrointestinal bleeding in two patients and liver dysfunction in one. The follow-up study showed no evidence of recurrence in any of the 13 patients up to 143 months after operation. In the operated patients, a small haemangioma was left in the liver of two for anatomical reasons, but the size of these tumours showed no change over 3 years. The remaining 20 patients had no specific treatment for haemangioma and were followed up for between 18 months and 8 years. Of these 20 patients, 19 had tumours of less than 5 cm in size. As there was no enlargement or rupture of haemangiomas of less than 5 cm, it was felt that these could be observed. Cases having a potential for exposure to trauma, rapid tumour growth, or displaying severe symptoms should be considered for surgical treatment; this can be justified due to low morbidity and no mortality.

Hepatic hemangiomas

There are 2 main and different forms of hepatic hemangiomas--those of early childhood and those of later adult life. The former, called infantile hepatic hemangioendothelioma, may be life-threatening due to arteriovenous shunting in extensive lesions resulting in cardiac failure. Although they may resolve spontaneously, if cardiac failure develops, they must be treated aggressively by arterial ligation or embolization. The adult form rarely gives rise to symptoms and should be left alone. Spontaneous rupture is extremely rare but may follow needle biopsy and it is in this group of patients that the occasional death has been reported. Biopsy should, therefore, be avoided. Symptomatic lesions should be resected in a specialized center where the morbidity and mortality of liver resection is acceptably very low.