MRI-detected tumor deposits in cT3 and cT4 rectal cancer following neoadjuvant chemoradiotherapy.

OBJECTIVES: To evaluate the identification of tumor deposits (TDs) and the prognostic significance of an MRI tumor regression grade for TDs in patients with rectal cancer treated with neoadjuvant chemoradiotherapy (nCRT). METHODS: Ninety-one patients with cT3 or cT4 rectal cancer who underwent surgery following nCRT between August 2014 and June 2020 were retrospectively analyzed. Changes in pre-nCRT MRI-detected TDs (mrTDs) were described as mrTD regression grade. The diagnostic performance of post-nCRT MRI-detected TDs (ymrTDs) was compared with histopathological reference standard. The correlation between ymrTDs, mrTD regression grade, and disease-free survival (DFS) was assessed. RESULTS: The sensitivity and specificity of ymrTDs were 88.00% and 89.39%, respectively. The area under the receiver operating characteristic curve was 0.887 (95% confidence interval [CI]: 0.803-0.944). The 3-year DFS of patients with positive ymrTDs was significantly lower than of the negative group (44.83% vs 82.73%, p < 0.001). The 3-year DFS was 33.33% for patients with poor regression of mrTDs following nCRT and 55.56% for those with moderate regression, compared to 69.23% in good responders and 83.97% in patients without mrTDs (p < 0.001). On multivariable Cox regression, mrTD regression grade was the only independent MRI factor associated with DFS (p = 0.042). CONCLUSIONS: Diagnostic performance of ymrTDs was moderate. The mrTD regression grade was independently correlated with DFS, which may have a prognostic implication for treatment and follow-up. CLINICAL RELEVANCE STATEMENT: Patients with poor regression of MRI-detected tumor deposits may benefit from more aggressive treatments, such as chemoradiation therapy plus induction or consolidation chemotherapy. KEY POINTS: • MRI provides a preoperative and noninvasive way to visualize tumor deposits (TDs) after neoadjuvant chemoradiotherapy (nCRT). • Post-nCRT MRI-detected TDs are a poor prognostic marker in cT3 and cT4 rectal cancer patients. • The regression of MRI-detected TDs after nCRT is associated with an improved disease-free survival.

A comparative study between deep learning and radiomics models in grading liver tumors using hepatobiliary phase contrast-enhanced MR images.

PURPOSE: To compare a deep learning model with a radiomics model in differentiating high-grade (LR-3, LR-4, LR-5) liver imaging reporting and data system (LI-RADS) liver tumors from low-grade (LR-1, LR-2) LI-RADS tumors based on the contrast-enhanced magnetic resonance images. METHODS: Magnetic resonance imaging scans of 361 suspected hepatocellular carcinoma patients were retrospectively reviewed. Lesion volume segmentation was manually performed by two radiologists, resulting in 426 lesions from the training set and 83 lesions from the test set. The radiomics model was constructed using a support vector machine (SVM) with pre-defined features, which was first selected using Chi-square test, followed by refining using binary least absolute shrinkage and selection operator (LASSO) regression. The deep learning model was established based on the DenseNet. Performance of the models was quantified by area under the receiver-operating characteristic curve (AUC), accuracy, sensitivity, specificity and F1-score. RESULTS: A set of 8 most informative features was selected from 1049 features to train the SVM classifier. The AUCs of the radiomics model were 0.857 (95% confidence interval [CI] 0.816-0.888) for the training set and 0.879 (95% CI 0.779-0.935) for the test set. The deep learning method achieved AUCs of 0.838 (95% CI 0.799-0.871) for the training set and 0.717 (95% CI 0.601-0.814) for the test set. The performance difference between these two models was assessed by t-test, which showed the results in both training and test sets were statistically significant. CONCLUSION: The deep learning based model can be trained end-to-end with little extra domain knowledge, while the radiomics model requires complex feature selection. However, this process makes the radiomics model achieve better performance in this study with smaller computational cost and more potential on model interpretability.

Percutaneous Gastrostomy Compared with Esophageal Stent Placement for the Treatment of Esophageal Cancer with Dysphagia.

PURPOSE: To compare the outcomes of self-expandable metal stent placement and percutaneous gastrostomy (PG) for the treatment of patients with esophageal cancer (EC) and dysphagia. MATERIALS AND METHODS: This retrospective observational study consisted of 113 patients with EC and dysphagia who underwent either stent placement (n = 47) or PG (n = 66) at a single center between June 2014 and June 2018. RESULTS: There were 63 men and 50 women, with a mean age of 76.5 years (standard deviation 4.9 years). The 2 groups had similar baseline characteristics, except that the PG group had a higher percentage of patients with cervical EC (22.7% vs 2.1%, P < .001). The PG group had better maintenance of nutritional status in terms of reduction in serum albumin level (P = .039) and weight loss (P = .041). Compared with the stent group, the PG group demonstrated a lower incidence of local severe pain (0% vs 21.3%, P < .001) and lower incidence of dislodgment of device (1.5% vs 19.1%, P = .002). The PG group demonstrated longer overall survival compared with the stent group for Stages II and III (201 vs 185 days, P = .034) and Stage IV (122 vs 86 days, P = .001). CONCLUSIONS: Compared with stent insertion, PG is associated with better maintenance of nutritional status, fewer complications, and better survival. Thus, PG may be the preferred choice for treating malnutrition in patients with EC and dysphagia.

Pelvic floor assessment using magnetic resonance imaging after vaginal delivery and elective caesarean delivery.

INTRODUCTION AND HYPOTHESIS: We aimed to compare quantitative static and dynamic magnetic resonance imaging (MRI) measurements of pelvic floor changes during postpartum recovery from 1 week to 6 months after different modes of delivery. METHODS: In this prospective study, 51 primiparous women (vaginal delivery group: 30 women; elective caesarean delivery group: 21 women) underwent static and dynamic MRI at 1 week, 6 weeks, 3 months, and 6 months postpartum to measure pelvic floor MRI values. Between-group differences in pelvic floor values at these time points were determined; subsequently, within-group comparisons according to time were performed. Analysis included independent samples t-tests and paired t-tests. RESULTS: The puborectal hiatus line (H line), muscular pelvic floor relaxation line (M line), bladder-pubococcygeal line (B-PCL), and uterus-pubococcygeal line (U-PCL) differed significantly between groups during the Valsalva manoeuvre at 1 week postpartum (p < 0.05). The H line, M line, and B-PCL values differed significantly between groups during the Valsalva manoeuvre at 6 weeks postpartum (p < 0.05). There were few significant between-group differences in pelvic floor values at 3 months and 6 months postpartum. In the vaginal delivery group, the differences in the H line and M line at 1 week, 6 weeks, and 3 months postpartum were significant (p < 0.001). In the elective caesarean delivery group, U-PCL differed significantly at 6 weeks compared to 1 week postpartum during the Valsalva manoeuvre (p < 0.05). CONCLUSIONS: Pelvic floor recovery primarily occurred during the early phase after delivery in both groups. Elective caesarean delivery had a non-significant protective effect on postpartum pelvic floor structure and function compared to vaginal delivery.

Unveiling the future: Advancements in MRI imaging for neurodegenerative disorders.

Neurodegenerative disorders represent a significant and growing global health challenge, necessitating continuous advancements in diagnostic tools for accurate and early detection. This work explores the recent progress in Magnetic Resonance Imaging (MRI) techniques and their application in the realm of neurodegenerative disorders. The introductory section provides a comprehensive overview of the study's background, significance, and objectives. Recognizing the current challenges associated with conventional MRI, the manuscript delves into advanced imaging techniques such as high-resolution structural imaging (HR-MRI), functional MRI (fMRI), diffusion tensor imaging (DTI), and positron emission tomography-MRI (PET-MRI) fusion. Each technique is critically examined regarding its potential to address theranostic limitations and contribute to a more nuanced understanding of the underlying pathology. A substantial portion of the work is dedicated to exploring the applications of advanced MRI in specific neurodegenerative disorders, including Parkinson's disease, Alzheimer's disease, Huntington's disease, and Amyotrophic Lateral Sclerosis (ALS). In addressing the future landscape, the manuscript examines technological advances, including the integration of machine learning and artificial intelligence in neuroimaging. The conclusion summarizes key findings, outlines implications for future research, and underscores the importance of these advancements in reshaping our understanding and approach to neurodegenerative disorders.

Round Window Niche Veil is Visible on High-Resolution Computed Tomography and a Predictor of Local Drug Efficacy to Inner Ear.

OBJECTIVES: To determine the morphologies and effect of the round window niche veil (RWNV) on local drug delivery efficacy and develop diagnostic criteria on high-resolution computed tomography (HRCT). METHODS: Patients diagnosed with otosclerosis, bilateral profound sensorineural hearing loss or vestibular schwannoma were enrolled from 2019 to 2022, receiving temporal bone HRCT scanning, and anatomic variations of RWMV were summarized intraoperative. For patients with vestibular schwannoma, 1 mL of dexamethasone solution (4 mg/mL) was administered via facial recess during operation, and samples of perilymph were collected to analyze. The diagnostic criteria of RWNV on HRCT were developed and verified. RESULTS: A total of 85 patients were enrolled. RWNV was observed in 54 cases intraoperatively with an incidence of 63.5% (95% CI, 52.9%-73.0%). The median perilymph concentrations were 4.86-fold higher in the group without RWNV than with RWNV (p < 0.0001). RWNV could be visualized on HRCT with a window width of 3500-4500 HU and a window level of 300-500 HU. The characteristic features were as follows: (1) a thin soft tissue shadow could be seen at the entrance of the round window niche (RWN); (2) it was visible in at least 2 consecutive layers along the upper margin of RWN from top to bottom; (3) it was discontinuous with the adjacent bone margin. The sensitivity and specificity of the diagnostic criteria were 77.8% and 93.6%, respectively. CONCLUSION: RWNV could reduce local dexamethasone diffusion efficacy to the inner ear, which could be diagnosed on HRCT and used as a predictor of local drug delivery efficacy to the inner ear. LEVEL OF EVIDENCE: 3 Laryngoscope, 134:1396-1402, 2024.

The usefulness of preoperative computed tomography and intraoperative neuromonitoring identification of the nonrecurrent inferior laryngeal nerve.

The objective of this study was to avoid the nonrecurrent inferior laryngeal nerve (NRLN) injury during surgery, we performed preoperative CT examinations to determine the variation in abnormal course of the right subclavian artery as an indictor of the presence of the NRLN and used intraoperative neuromonitoring (IONM) to identify nerve. Preoperative thyroid CT examinations were performed in 783 thyroid surgery patients. The imaging characteristics that suggested the presence of the NRLN were the following: (1) the arteria lusoria arising from the dorsal side of the aortic arch and passing through the trachea and esophagus posteriorly, and the CT image showing the characteristic "hook-like" morphology; (2) that the arteria lusoria imaging could be observed posteriorly to the trachea and esophagus; and (3) that the arteria lusoria traveled transversely from the rear of the right common carotid artery to the right subaxillary region. IONM has been applied to localize and identify NRLN. The brachiocephalic trunk was shown in 779 cases and not in the remaining four cases (0.5 %, 4/783), and these four were assumed to have the arteria lusoria. The separation point and path of the NRLNs were localized and identified precisely with IONM. The NRLN was observed during all surgeries. These four cases did not exhibit hoarseness after surgery. In conclusion, understanding of the course variations of the right subclavian artery using a preoperative CT examination provides an indicator of the presence of a NRLN. Combining these evaluation methods with IONM can avoid NRLN injury.

Dictionary learning LASSO for feature selection with application to hepatocellular carcinoma grading using contrast enhanced magnetic resonance imaging.

Introduction: The successful use of machine learning (ML) for medical diagnostic purposes has prompted myriad applications in cancer image analysis. Particularly for hepatocellular carcinoma (HCC) grading, there has been a surge of interest in ML-based selection of the discriminative features from high-dimensional magnetic resonance imaging (MRI) radiomics data. As one of the most commonly used ML-based selection methods, the least absolute shrinkage and selection operator (LASSO) has high discriminative power of the essential feature based on linear representation between input features and output labels. However, most LASSO methods directly explore the original training data rather than effectively exploiting the most informative features of radiomics data for HCC grading. To overcome this limitation, this study marks the first attempt to propose a feature selection method based on LASSO with dictionary learning, where a dictionary is learned from the training features, using the Fisher ratio to maximize the discriminative information in the feature. Methods: This study proposes a LASSO method with dictionary learning to ensure the accuracy and discrimination of feature selection. Specifically, based on the Fisher ratio score, each radiomic feature is classified into two groups: the high-information and the low-information group. Then, a dictionary is learned through an optimal mapping matrix to enhance the high-information part and suppress the low discriminative information for the task of HCC grading. Finally, we select the most discrimination features according to the LASSO coefficients based on the learned dictionary. Results and discussion: The experimental results based on two classifiers (KNN and SVM) showed that the proposed method yielded accuracy gains, compared favorably with another 5 state-of-the-practice feature selection methods.

Glutathione and Esterase Dual-Responsive Smart Nano-drug Delivery System Capable of Breaking the Redox Balance for Enhanced Tumor Therapy.

Conventional chemotherapy usually fails to achieve its intended effect because of the poor water solubility, poor tumor selectivity, and low tumor accumulation of chemotherapy drugs. The systemic toxicity of chemotherapy agents is also a problem that cannot be ignored. It is expected that smart nano-drug delivery systems that are able to respond to tumor microenvironments will provide better therapeutic outcomes with decreased side effects of chemotherapeutics. Nano-drug delivery systems capable of breaking the redox balance can also increase the sensitivity of tumor cells to chemotherapeutics. In this study, using polymer-containing disulfide bonds, ester bonds, and d-α-tocopherol polyethylene glycol succinate (TPGS), which can amplify reactive oxygen species (ROS) in tumor cells, we have successfully prepared a smart glutathione (GSH) and esterase dual-responsive nano-drug delivery system (DTX@PAMBE-SS-TPGS NPs) with the ability to deplete GSH as well as amplify ROS and effectively release an encapsulated chemotherapy drug (DTX) in tumor cells. The potential of DTX@PAMBE-SS-TPGS NPs for enhanced antitumor effects was thoroughly evaluated using in vitro as well as in vivo experiments. Our research offers a promising strategy for maximizing the efficacy of tumor therapy.

Meta-analysis assessing the effectiveness of SGLT2i+GLP1RA combination therapy versus monotherapy on cardiovascular and cerebrovascular outcomes in diabetic patients.

Relevant meta-analyses have confirmed the cardiovascular and renal benefits of sodium-glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP1RA) among patients with type 2 diabetes (T2D) and/or cardiorenal disease. However, it is not established whether the combination therapy of SGLT2i and GLP1RA will yield an additive benefit on cardiorenal endpoints. Lopez and colleagues recently did a cohort study (Lopez et al., Am. J. Cardiol., 2022, 181, 87-93) and aimed to address this issue. However, their findings are not consistent with those of previous studies. To confirm Lopez et al.'s findings (Lopez et al., Am. J. Cardiol., 2022, 181, 87-93) and address the aforementioned inconsistencies, we conducted a meta-analysis based on relevant studies. Our meta-analysis identified that SGLT2i + GLP1RA combination therapy was significantly associated with the reduced risks of cardiovascular/cerebrovascular atherosclerotic, heart failure-associated, and death outcomes compared with SGLT2i/GLP1RA monotherapy. These might support this combination therapy used for better reducing cardiovascular and death events in T2D patients, especially in those with high or very high cardiovascular risk. This is a commentary on a previous article (Lopez et al.'s study (Lopez et al., Am. J. Cardiol., 2022, 181, 87-93)) published outside of Frontiers. Therefore, we submitted this manuscript as an Opinion article, as suggested in the Author Guidelines.

Ultrasonic misdiagnosis of giant pediatric testicular yolk sac tumor: A case report and literature review.

Background: Yolk sac tumor is the most common malignant nonseminomatous germ-cell tumor in children characterized by elevated level of α-fetoprotein (AFP), accounting for 70%-80% of all cases. However, giant yolk sac tumors that involve the entire testicle may be misdiagnosed by color Doppler ultrasonography as orchitis. Therefore, we described a case of a 2-year-old pediatric patient with a giant testicular yolk sac tumor that was misdiagnosed by ultrasonography as orchitis, in order to evaluate the role of measuring AFP levels in the initial diagnosis to aid in the accuracy of the definitive diagnosis of testicular yolk sac tumor. Case presentation: A 2-year-old boy received outpatient visits for unintentional swelling of the right scrotum for 7 days. Physical examination showed a rubbery swelling of the right scrotum with rejective touch. Then, the patient underwent perineal color Doppler ultrasonography in outpatient visits. The result showed a right testicle size of 29 mm × 22 mm × 20 mm with heterogeneous echogenicity and abundant blood flow, supporting the initial diagnosis of orchitis. However, the initial surgeon was skeptical of the ultrasonography diagnosis. Thus, the patient was admitted to the Department of Andrology on day 2 for further serological and imaging examination. The serum AFP level on day 3 was 323.77 ng/ml. The results of CT and MRI showed a giant tumor of the right testis (26 mm × 21 mm × 29.6 mm) with multiple lymphoid hyperplasia in the inguinal region bilaterally. The patient received radical orchidectomy without lymph node dissection on day 9. The results of postoperative pathological examination confirmed giant testicular yolk sac tumor (T1N0M0S1, Stage Is) and was positive for AFP and SALL4 in immunohistochemistry staining. The patient received three courses of bleomycin-etoposide-cisplatin chemotherapy in the Department of Pediatrics after multidisciplinary team meeting on postoperative days 14, 37, and 58, respectively. During chemotherapy and follow-up, the patient's AFP and lactate dehydrogenase levels continued to decline, and eventually remained within normal range on postoperative day 84. Conclusion: Measuring the AFP level was necessary for initial diagnosis and follow-up in pediatric cases of testicular enlargement. Radical orchidectomy combined with postoperative bleomycin-etoposide-cisplatin adjuvant chemotherapy was an effective treatment strategy for pediatric giant testicular yolk sac tumors.

Video-based education improves the image quality of diagnostic percutaneous cerebral angiography among elderly patients.

OBJECTIVE: Digital subtraction angiography (DSA) is considered the gold standard for cerebral vasculature observation and is increasingly applied among the elderly population. The aim of this study is to determine whether the use of a video-based education system can improve the image quality of percutaneous cerebral angiography. METHOD: This study is a single-blinded prospective cohort trial. One hundred and sixty patients (≥65 years old) were enrolled in this study. Eighty patients were provided with video-based education as intervention. Eighty age-matched controls only received regular education. The DSA image quality was assessed between control and intervention groups. It was rated by two readers on a 5-point scale, independently. RESULTS: No differences were found between control and intervention groups in baseline characteristics (P > 0.05). The mean overall image quality was significantly higher in patients receiving video-based education than in controls (P < 0.05), and the same trends were found in the respective assessment of each artery (left and right carotid/vertebral artery; P < 0.05). Moreover, the operation time and radiation doses were quite comparable between the two groups (P > 0.05). CONCLUSIONS: This study indicated that video-based education helps elderly patients to acquire improved DSA image quality. It encourages the application of this approach in practice.

[Adenovirus mediated gene transfer of tyrosinase gene on HepG2 cell by magnetic resonance imaging].

OBJECTIVE: To evaluate the transfect results of recombinant adenovirus vector carrying tyrosinase gene (Ad-tyr) in vitro by magnetic resonance imaging (MRI) after the Ad-tyr was transfected into HepG2 cell. METHODS: The Ad-tyr which carried the full-length cDNA of tyrosinase gene was transfected into HepG2 cell. The transfected cells were scan by MRI sequences of T1 weighted image (T1WI) , T2 weighted image (T2WI) , and short time inversion recovery (STIR) to observe the MRI signals of expressed melanin. Masson-Fontana staining was performed to search for melanin granules in transfected cells. Real-time PCR method was used to search for cDNA of tyrosinase gene. RESULTS: Ad-tyr was transfected into HepG2 cells and synthesized a large amount of melanin inside. The synthesized melanin of 1 x 10(6) cells which had been transfected by Ad-tyr with the 50, 150, and 300 multiplicity of infection separately were all sufficient to be detected by MRI and showed high signals in MRI T1WI, T2WI, and STIR sequences. The signal intensities of MRI were positively correlated to the amounts of transfected Ad-tyr. The melanin granules were found in HepG2 cells in Masson-Fontana staining. The cDNA amount of tyrosinase gene in transfected HepG2 cells, which was detected by real-time PCR, was remarkably higher than that in nontransfected cells. CONCLUSION: The synthesized melanin of HepG2 cells, which controlled by expression of exogenous gene, can be detected by MRI, indicating that the adenovirus vector can efficiently carry the tyrosinase gene into HepG2 cells.

[Value of magnetic resonance imaging in displaying the parotid gland segments of the facial nerve].

OBJECTIVE: To evaluate the value of magnetic resonance imaging (MRI) in displaying the parotid gland segments of the facial nerve. METHODS: Sixteen volunteers (9 males and 7 females) and 132 surgically confirmed patients with parotid tumors locating in the deep or shallow lobe of the parotid gland (including 89 with benign and 43 with malignant tumors) underwent MRI using T1WI and T2WI. The transverse images were obtained with the plane tilted 35 degrees to the foot, and the coronal images were acquired using conventional scanning. RESULTS: On transverse T1WI, the parotid gland segments of the facial nerve displayed low signal with arc-shaped curve in the cross-section, showing a symmetrical dot-like low signal in the coronal plane. The facial nerve in 63% of the patients with parotid tumors in the cross-section could be displayed, but in the coronal plane the proportion reached 83%. MRI could accurately reveal the position of the parotid tumors in the deep or shallow lobe of the parotid gland. CONCLUSION: MRI can show the major portion of the parotid gland segments of the facial nerve and has important value in locating the parotid tumors and displaying the relationship between the tumor and facial nerve.

[Magnetic resonance imaging for diagnosis of parotid malignant tumors and the pathological basis].

OBJECTIVE: To investigate magnetic resonance imaging (MRI) features of parotid malignant tumors and study their pathological basis. METHODS: Forty-seven patients with parotid malignant tumors confirmed by surgery (41 patients) or biopsy (6 patients) were enrolled in this study. A comparative analysis was conducted of the pathological and MRI findings in 30 patients with the entire lesions available. Each of the MRI features was analyzed retrospectively and the typical MRI findings of common parotid malignant tumors were summarized. RESULTS: MRI allowed accurate diagnosis of parotid malignant tumors. Four patients with low-grade mucoepidermoid carcinoma showed well-defined tumor margin and were difficult to distinguish from benign tumors. Six patients with high-grade mucoepidermoid carcinoma had obscure margin of the tumor which easily underwent necrosis with liable lymph node involvement. The 8 cases of adenoid cystic carcinoma was characterized by extensive invasion surrounding the parotid gland. Most of 8 cases of malignant pleomorphic adenoma still showed high and heterogeneous signal on T2WI, with irregular shape and poorly defined margin. Nine cases of lymphoma all had secondary lesions characterized by extensive involvement and presence of multiple nodules. The 4 cases of acinic cell carcinoma showed either regular or irregular tumor morphology, presenting with high signal intensity on T1WI and T(2)WI. CONCLUSION: MRI is an important modality for the diagnosis of parotid malignant tumors. Most of the common parotid malignant tumors have characteristic MRI and pathological features, which make possible their differential diagnosis.

[MRI performance and diagnosis of schwannomas from cranial nerves in posterior cranial fossae].

BACKGROUND & OBJECTIVE: Cranial nerve schwannomas originate frequently in posterior cranial fossae and have various and complex MRI performances, some of which are still not well known. This study was to explore MRI performances and features of schwannomas from cranial nerves in posterior cranial fossae. METHODS: The MRI performances of 75 cases of schwannoma from cranial nerves in posterior cranial fossae, including trigeminal (n=9), facial (n=1), acoustic (n=53), 9th-11th (n=9) and hypoglossal (n=3) schwannomas, confirmed by surgical and pathologic findings, were analyzed retrospectively. RESULTS: Most of schwannomas in posterior cranial fossae were solid-cystic lesions when their sizes were larger than 1.5 cm in diameter. Small lesions (less than 1.5 cm in diameter) may be completely solid, which were closely related to cranial nerves. On T1WI, the solid part of tumor appeared iso- or slightly hypointense, while cystic part was hypointense. On T2WI, solid part appeared high or slightly high signal intensity, but cystic part appeared very high signal intensity. On contrast-enhanced T1WI, there was obvious enhancement in the solid part, but not in the cystic part. Some typical signs were very useful to infer tumor origin, such as, dumbbell-shaped trigeminal schwannoma extended across the middle and posterior cranial fossa, enlargement of internal auditory canal, widened jugular foramen and hypoglossal foramen caused by acoustic schwannoma, the 9th-11th shcwannoma, and hypoglossal schwannoma, respectively. The correct ratio for qualitative diagnosis of schwannoma was 92% using MRI, but the incorrect ratio for identifying the nerve of tumor origin was 8.7%. CONCLUSION: MRI is a good method in qualitative diagnosis of schwannoma and identifying cranial nerves of tumor origin in posterior cranial fossae.

[Types of liver metastases on MR imaging and characteristics of uncommon types].

BACKGROUND & OBJECTIVE: Although liver metastases are very common, some of them would be misdiagnosed because of their specific appearance on imaging. This study was to summarize the types of liver metastases according to the magnetic resonance imaging (MRI) appearance, and explore the characteristics of uncommon types to improve the accuracy of diagnosis. METHODS: A total of 174 consecutive patients with liver metastases were screened from the patients with diagnosis or suspicion of liver metastases according to plate or enhanced MRI scan. Liver metastases were confirmed through pathologic examination, medical imaging, or follow-up. Lesions were categorized as nodular, segment, and diffusion classes by shape, and categorized as common, mimicking hemangioma, and mimicking cyst patterns according to the characteristics of signal intensity. RESULTS: Common primary tumors of liver metastases were colon and rectum carcinomas (31.0%). According to the shape of lesions, the proportions of nodular, segment, and diffusion classes were 96.6%, 2.3%, and 1.1%, respectively. According to the characteristics of signal intensity, the proportions of common, mimicking hemangioma, and mimicking cyst patterns were 87.9%, 4.0%, and 8.1%, respectively. CONCLUSIONS: Most liver metastases are nodular in shape, and present as common pattern, but some uncommon types also exist. MRI scan and clinical follow-up could increase the accuracy of diagnosis.

[Magnetic resonance imaging (MRI) in evaluation of tyrosinase gene expression in HepG2 cell].

BACKGROUND & OBJECTIVE: Gene therapy is the frontier of life science. There is no perfect method to evaluate gene expression without invasion at present. Medical imaging connecting with molecular biology might be helpful; however, the technology is just on the horizon. The authors conducted this study in vitro by transferring reporter tyrosinase gene into HepG2 cell to apply magnetic resonance imaging (MR) for evaluating gene expression. METHODS: The plasmid of pcDNA3tyr carried the full-length cDNA of tyrosinase gene was transfected into HepG2 cell by lipofectin, its property of synthesizing melanin was used to produce high signal in T1WI MR image and then to evaluate gene expression. Further identification were performed with searching melanin granules by Fontana staining and searching cDNA of tyrosinase gene using reverse transcription-polymerase chain reaction (RT-PCR). RESULTS: (1)The melanin granules were found in HepG2 cell using Fontana staining. (2)The cDNA fragment of tyrosinase gene was detectable in transferred HepG2 cell by RT-PCR. (3)Plasmid of pcDNA3tyr was transfected into HepG2 cell and synthesized a large amount of melanin in HepG2 cell; the synthetic melanin appeared high signal in T1WI MR imaging as same as natural melanin and was enough to be detected by MR. And further, the signal intensity was positively related to the amount of transferred plasmid. CONCLUSION: The fact that synthetic melanin of HepG2 cell can be detected by MR demonstrates that medical imaging connecting with molecular biology can be used to evaluate the result of gene expression in vitro.

[MRI performance and diagnosis of meningioma--a report of 126 cases].

BACKGROUND & OBJECTIVE: In MRI study of meningioma, it was lack of large group of patients to evaluate MRI qualitative diagnosis, and no consensus had been achieved concerning problems such as peritumoral edema in meningioma. This study was to summarize main clues for diagnosing meningioma through analyzing MRI performance of 126 patients with meningioma. METHODS: Among 126 patients with meningioma, 32 were syncytial, 35 were fibroblastic, 24 were psammomatous, 9 were angioblastic, 18 were transitional, 3 were papillary, and 5 were malignant. All patients were scanned with T1, T2-weighted imaging (T1WI, T2WI), and contrast-enhanced T1WI. RESULTS: Convexity of brain was more likely to be involved, among 126 cases of meningioma, 45 (35.7%) tumors located at convexity of brain. The size of tumor ranged from 1.4 to 9.9 cm. Eighty-one percent of tumors were round or oval in shape. Isointensity or slight hyperintensity of T2WI signals detected in 70.6% patients. The rates of tail sign, and pseudo-capsule were 62.7%, and 49.2%. Extruding sign of brain parenchyma was observed in 83.8% (57/68)of patients with tumor size of > 4 cm. Significantly even and increasing sign in contrast-enhanced T1WI were observed in 104 patients (82.5%). Peritumoral edema occurred in 57 patients (45.1%), and related to tumor size. Other rare signs included cystic changes, bleeding, calcification, osteal changes, and introtumoral vessel symptoms. The correct rate of diagnosis was 95.2%. CONCLUSIONS: MRI performances of meningioma are various. Judgment of extra-brain tumor, typical T2WI signals, tail sign, and significantly even and increasing sign are key factors for diagnosing meningioma.