MR-guided interstitial laser thermotherapy of colorectal liver metastases: efficiency, safety and patient survival.

PURPOSE: Evaluation of MR-guided interstitial laser thermotherapy (ILT) of colorectal liver metastases under consideration of efficacy, safety and patient survival. MATERIALS AND METHODS: Sixty-six inoperable patients with a total of 117 colorectal liver metastases were treated with MR-guided laser therapy in 96 sessions. 40.9% of patients presented metastases from rectum carcinoma, 30.3% from sigmoid carcinoma and 28.8% from colon carcinoma. Inclusion criteria were < or =5 metastases < or =5 cm in greatest diameter and no extrahepatic tumor spread. Internally water-cooled 9F power-laser-applicators were placed under CT-fluoroscopy. For MR-guided ILT, a 1064 nm Nd-YAG-lasers with a beam divider with multi applicator technique was used. The energy applied was 10 watt per centimeter diffusor length, with the diffusor length ranging from 20 to 40 mm. The mean duration of the energy application was 23 minutes (range: 15 - 37 minutes). The endpoint of the laser ablation was defined as the absence of hyperintense tumor tissue in the continuously monitored T2-w fat saturated gradient-echo sequences. Follow-up included contrast-enhanced MRI using T1- and T2-weighted spin-echo and gradient-echo sequences every three months after treatment. Survival times were calculated using the Kaplan-Meier method. RESULTS: The median follow-up was 8.7 months (mean 11.8; standard deviation 9.9; range 1 to 36). The overall median progression free survival was 6.1 months (range, 0.3 to 27+ months). Median survival was 23 months (95% CI, 17-29 months). The rate of major complications was 2.1% (n = 2) and peri-procedural mortality (30 days) was 3% (n = 2). After 3, 6, 9, and 12 months, local tumor control was 98.3%, 91.4%, 76.1%, and 69.4%, respectively. In no patient metastatic deposits along the catheter access route were found. CONCLUSIONS: In patients with colorectal liver metastases, interstitial laser thermotherapy is an effective and safe therapeutic option and therefore suitable not only in palliative situations.

Differentiation between malignant and benign solitary pulmonary nodules with proton density weighted and ECG-gated magnetic resonance imaging.

OBJECTIVE: To estimate performance of MRI for differentiating malignant from benign solitary pulmonary nodules (SPN) using morphological characteristics. MATERIAL AND METHODS: MRI in 46 patients with SPN (mean diameter: 19 mm) was carried out on 1.0 Tesla scanner using ECG-gated, gradient echo sequence. Morphological signs of SPN were determined and compared with previously performed helical-CT, where final diagnosis served as reference with 52% frequency of malignancy. Furthermore, three observers evaluated all images. RESULTS: Significant differences between the two groups were found for nodules shape, margin, inhomogeneity and the vessel-sign in MRI, nodules shape, margin, the vessel-sign, and presence of spicules in CT. Using these signs, AUC were 0.746 for MRI and 0.765 for CT. The mean sensitivity, specificity, and accuracy of observers for MRI/CT were 89%/95%, 42%/41%, 66%/68%, respectively. CONCLUSIONS: Despite discrepancies in morphologic appearance, no significant difference of accuracy between MRI and CT was determined. Further investigations are necessary to demonstrate the clinical use in combination with functional parameters, establishing MRI as a comprehensive diagnostic modality for SPN.

Determination of the optimal MRI sequence for the detection of malignant lung nodules.

OBJECTIVE: For staging, follow-up and even screening (www.screening.info) an "all-in-one" imaging examination is desirable. In the concept of whole body MRI, lung imaging prevails as the weakest link. The purpose of our study was to determine the optimal MRI sequences for the detection of malignant lung nodules. PATIENTS AND METHODS: On the basis of 6 lung cancer, 46 metastases and one tuberculoma in 13 patients eight MRI sequences--HASTE, IR-HASTE, fat saturated TrueFISP, STIR, VIBEipat = 2, and contrast-enhanced (CE) VIBE (with ipat = 2, 0, 4) performed with parallel imaging and 12 matrix coil elements--were compared in terms of contrast-to-noise ratio (CNR) and quality in the visualization of the lung nodules using multidetector CT as standard of reference. The parameters of the sequences were pragmatically selected to minimize the imaging time to allow for imaging the entire lung within one breathold interval. RESULTS: The STIR sequence was found to be the best for detecting malignant lung nodules (p<0.01) followed by the FS TrueFISP, CE VIBE subsetipat = 0, CE VIBE subsetipat = 2, IR-HASTE, HASTE, CE VIBE subsetipat = 4, and VIBE. The STIR sequence visualized malignant nodules down to 2 mm in size and did not display the 19 mm tuberculoma. CONCLUSION: The STIR sequence should be included in future studies investigating if MRI can compete with CT in the early identification (detection and classification) of malignant lung nodules.

Image fusion of CT and MRI for the visualization of the auditory and vestibular system.

The aim of this study was to perform a realistic visualization of the auditory and vestibular system using volume data sets from high-resolution computed tomography (HR-CT) and high-resolution magnetic resonance imaging (HR-MRI). - In 10 patients with conductive and/or sensorineural hearing loss, vertigo and tinnitus, HR-CT and HR-MRI of the petrous bone were performed consecutively. CT was performed with a 16-slice computed tomography scanner using a high spatial resolution. MRI was performed with a 3.0 Tesla scanner using a three-dimensional-constructive interference in steady state (3D-CISS) gradient-echo, and T2-weighted, unenhanced and gadolinium (GD)-enhanced T1-weighted turbo spin-echo sequences. The middle ear structures were interactively segmented and visualized with a color-coded shaded-surface rendering method using the HR-CT volume data sets. The inner ear structures were interactively segmented and visualized with a color-coded shaded-surface rendering method using the high-resolution 3D-CISS MRI volume data sets. Finally, both shaded-surface rendered models were superimposed semi-automatically using a commercial available software program to visualize the auditory and vestibular system. - The representation of the middle and inner ear structures with image fusion of HR-CT and HR-MRI takes advantage of both the high bony contrast of HR-CT and the high soft tissue contrast discrimination and sensitivity to fluids of HR-MRI, as well as the high spatial resolution of both modalities. In comparison to the fused axial CT/MRI, the images of 3D CT/MRI fusion facilitates a clear representation and better spatial orientation. - The middle and inner ear consists of bony structures, soft tissue structures and fluid-filled spaces. For this reason, the image fusion of volume data sets from HR-CT and HR-MRI allowed an optimized and realistic visualization of the auditory and vestibular system.

[Detection of pulmonary nodules with breath-hold magnetic resonance imaging in comparison with computed tomography]. / Detektion von Lungenrundherden mit der Magnetresonanztomographie in Atemanhaltetechnik im Vergleich zur Spiral-Computertomographie.

PURPOSE: Evaluation of sensitivity and false positive findings of two fast MRI sequences for the detection of pulmonary nodules in comparison with spiral CT by two independent observers. MATERIALS AND METHODS: All 30 enrolled patients had a spiral CT or MSCT as base line study. MRI was performed with a 1.5 T MR scanner (Sonata, Siemens) using a transverse 3D gradient echo sequence (3D-GRE: TR/TE/flip = 2.9 ms/1.1 ms/5 degrees ) and a half-Fourier single-shot fast spin-echo sequence (HASTE: TR/TE/flip = 800/25/150 degrees ) acquired in three planes. A separate analysis for both sequences was carried out prospectively by two independent readers (A and B) with different experience regarding pulmonary MRI. Additionally, a retrospective reading with knowledge of the CT scans was done. Results were calculated for all lesions and for lesions larger than 4 mm. RESULTS: The sensitivities were 73 %, 70 % and 84 % for the 3D-GRE sequence (reader A, reader B, retrospective reading) and 65 %, 68 % and 81 % for the HASTE sequence. For lesions larger than 4 mm, the sensitivities were 93 %, 89 %, 96 % for the 3D-GRE sequence and 85 %, 85 %, 96 % for the HASTE sequence. The rate of false positive findings depended on the reader's experience, but was generally lower for the 3D-GRE sequence with 2 and 16 (reader A and B) false positive nodules compared to 4 and 40 false positive findings for the HASTE sequence. The 3D-GRE sequence was more accurate for both readers (reader A: p = 0.08, reader B: p = 0.00003). CONCLUSION: The sensitivity of MRI for the detection of lung nodules was only acceptable for lesions larger than 4 mm. The 3D-GRE sequence is superior to the HASTE sequence due to the reduced amount of false positive findings with comparable sensitivity.

Whole-body imaging with PET/MRI.

Whole-body positron emission tomography (PET) scanning with the radiolabeled glucose analogue 2-[fluorine-18]-fluoro-2-deoxy-D-glucose ( superset 18 F-FDG) can identify areas of cancerous involvement and distinguish malignant from benign lesions and therefore, plays an important role in the diagnosis and management of patients with cancer. PET facilitates the evaluation of metabolic and molecular characteristics of a wide variety of cancers, but it is limited in its ability to visualize anatomical structures. Whole-body magnetic resonance imaging (MRI) is a promising diagnostic modality for the diagnosis and management of patients with cancer, because of its high anatomical resolution. Whole-body PET and whole-body MRI allow to evaluate both the primary tumor and for the presence of metastasis at the same time. The combination of these two excellent diagnostic imaging modalities into a single scanner offers several advantages in comparison to PET and MRI alone. A hybrid PET/MRI facilitates the accurate registration of metabolic and molecular aspects of the diseases with exact correlation to anatomical findings, improving the diagnostic value in identifying and characterizing of malignancies and tumor staging. Thus, hybrid PET/MRI could be a very important diagnostic imaging modality in oncological applications in the decades to come, and possibly for use in cancer screening and cardiac imaging.

FDG PET and other imaging modalities in the primary diagnosis of suspicious breast lesions.

Mammography is the primary imaging modality for screening of breast cancer and evaluation of breast lesions (T staging). Ultrasonography is an adjunctive tool for mammographically suspicious lesions, in patients with mastopathy and as guidance for reliable histological diagnosis with percutaneous biopsy. Dynamic enhanced magnetic resonance mammography (MRM) has a high sensitivity for the detection of breast cancer, but also a high false positive diagnosis rate. In the literature, MRM is reported to have a sensitivity of 86-96%, a specificity of 64-91%, an accuracy of 79-93%, a positive predictive value (PPV) of 77-92% and a negative predictive value (NPV) of 75-94%. In unclarified cases, metabolic imaging using fluorine-18 fluorodeoxyglucose positron emission tomography (FDG PET) can be performed. In the literature, FDG PET is reported to have a sensitivity of 64-96%, a specificity of 73-100%, an accuracy of 70-97%, a PPV of 81-100% and an NPV of 52-89%. Furthermore, PET or PET/CT using FDG has an important role in the assessment of N and M staging of breast cancer, the prediction of tumour response in patients with locally advanced breast cancer receiving neoadjuvant chemotherapy, and the differentiation of scar and cancer recurrence. Other functional radionuclide-based diagnostic tools, such as scintimammography with sestamibi, peptide scintigraphy or immunoscintigraphy, have a lower accuracy than FDG PET and, therefore, are appropriate only for exceptional indications.

MR visualization of the inner ear structures: comparison of 1.5 Tesla and 3 Tesla images.

AIM OF THE STUDY: To compare high resolution MRI examinations of inner ear structures at 1.5 T and at 3 T. METHOD: Temporal bones were measured bilaterally in 3 healthy volunteers in a 1.5 T and in a 3 T MR-scanner using the respective one channel head coil (quadrature detection) of the manufacturer. The same steady-state gradient echo sequence (3D-CISS) was employed at a voxel size of 0.4 x 0.4 x 0.4 mm(3). The signal-to-noise ratio (SNR) was determined quantitatively. RESULTS: An SNR of 8 could be achieved for the measurements at 3 T in 7:37 min. The SNR at 3 T was, on average, a factor of 1.34 higher than that at 1.5 T despite the fact that the excitation angle had to be drastically reduced (alpha = 42 degrees instead of alpha = 70 degrees at 1.5 T) due to the limit of the specific absorption rate (SAR). DISCUSSION: The MR representation of the inner ear is clearly improved at 3 T. To obtain the same SNR at 1.5 T approximately the double measuring time would be required, connected with reduced patient comfort and an increased risk for a displacement of the head during the high resolution measurement.

Assessment of peripheral arterial occlusive disease: comparison of multislice-CT angiography (MS-CTA) and intraarterial digital subtraction angiography (IA-DSA).

PURPOSE: The aim of the study was to assess the arterial vascular system of the lower extremities in patients with peripheral arterial occlusive disease using Multislice-CT angiography (MS-CTA) and to compare the results with the standard of reference, intraarterial digital subtraction angiography (IA-DSA). MATERIALS AND METHODS: MS-CTA and IA-DSA of the lower extremities were carried out on 23 patients with peripheral arterial occlusive disease (Fontain Stage IIb: 18, III: 3, IV: 2). MS-CTA comprised a 4 x 2.5 mm collimation, 15 mm table feed/rotation, 0.5 sec rotation time and 3 mm slice thickness (1.5 mm reconstruction increment). Delay time was determined by bolus tracking. 150 ml of contrast media were injected intravenously at a flow rate of 3 ml/sec. Maximum intensity projection (MIP) reformations were performed using a semi-automatic vessel tracking program. MS-CTA (axial and MIP-reformatted images) and IA-DSA were reviewed by two radiologists. The grade of vascular stenosis as well as occlusion were rated on a scale of 0 to 3 (0=0-50% stenosis, 1 = 51-75% stenosis, 2 = 76-99% stenosis, 3 = occlusion). RESULTS: For MS-CTA, the mean delay time was 30.2 s (23-40 s), mean scan time was 37.4 (33-42 s). Data analysis was based upon a total of 1136 vascular segments for both methods (568 each). A comparison of all the evaluated segments in both techniques revealed a MS-CTA / IA-DSA 86.3% match. Out of 442 segments proximal the trifurcation, 386 were correctly assessed in MS-CTA (87.3%) and distal the trifurcation, 101 out of 126 segments were rated correctly (80.2%). In MS-CTA, an overall confidence interval of 95% can be achieved in 83.2% to 89.0% for correctly rated stenosis grading. CONCLUSIONS: In patients with peripheral arterial occlusive disease, MS-CTA of the lower extremity is a promising minimal-invasive method for detection of relevant arterial stenoses. However, the technique was limited to routine diagnostic purposes due to severe calcifications and time consuming reconstruction procedures.

[Imaging diagnosis of solitary pulmonary nodules on an open low-field MRI system--comparison of two MR sequences with spiral CT]. / Bildgebende Diagnostik solitärer Lungenrundherde am offenen Niederfeld-MRT--Ein Vergleich zweier MR-Sequenzen mit der Spiral-CT.

OBJECTIVE: To evaluate and compare two fast gradient-echo sequences (GRE) concerning the visualization of solitary pulmonary nodules with an open low-field MRI system in comparison to computed tomography. MATERIALS AND METHODS: Fourteen patients with solitary pulmonary nodules detected by spiral CT ranging in size from 6 mm to 42 mm (mean 20 mm) underwent MRI on an open 0.2 T scanner using a spoiled 2D GRE (2D FLASH; TR/ TE/Flip = 100 ms/7.5 ms/30 degrees ) and a totally refocused 2D steady-state GRE (True-FISP; TR/TE/FA = 7.3 ms/3.5 ms/80 degrees ). The image quality concerning artifacts (by flow, breathing and susceptibility) and the morphologic characteristics of the nodules were scored and compared with CT by two independent radiologists. The diameters of the nodules measured by MRI were compared with CT measurements. The sequences were also evaluated with regard to the signal-to-noise ratio (SNR) of the lesion. RESULTS: All lesions were detected with the 2D FLASH sequence. True-FISP failed to find a granuloma with a size of 6 mm. The 2D FLASH was rated significantly superior to true FISP concerning image quality artifacts by susceptibility as well as concerning to CT the presentation of nodule characteristics. In MR images, the size of lesions was significantly smaller than in CT images for both sequences: for 2D FLASH the mean difference was 0.9 mm and for true FISP 2.6 mm. The SNR of the nodules was significantly higher for the 2D FLASH than for the true FISP. CONCLUSION: In low field MRI, the 2D FLASH sequence is superior to the 2D true FISP sequence in imaging of pulmonary nodules. With the 2D FLASH sequence nodules of 6 mm or larger in size can be visualized.

[Hybrid 3-D rendering of the thorax and surface-based virtual bronchoscopy in surgical and interventional therapy control]. / Hybride 3D-Visualisierung des Thorax und oberflächenbasierte virtuelle Bronchoskopie in der operativen und interventionellen Therapiekontrolle.

PURPOSE: The aim of this study was to demonstrate the possibilities of a hybrid rendering method, the combination of a color-coded surface and volume rendering method, with the feasibility of performing surface-based virtual endoscopy with different representation models in the operative and interventional therapy control of the chest. MATERIAL AND METHOD: In 6 consecutive patients with partial lung resection (n = 2) and lung transplantation (n = 4) a thin-section spiral computed tomography of the chest was performed. The tracheobronchial system and the introduced metallic stents were visualized using a color-coded surface rendering method. The remaining thoracic structures were visualized using a volume rendering method. For virtual bronchoscopy, the tracheobronchial system was visualized using a triangle surface model, a shaded-surface model and a transparent shaded-surface model. RESULTS: The hybrid 3D visualization uses the advantages of both the color-coded surface and volume rendering methods and facilitates a clear representation of the tracheobronchial system and the complex topographical relationship of morphological and pathological changes without loss of diagnostic information. Performing virtual bronchoscopy with the transparent shaded-surface model facilitates a reasonable to optimal, simultaneous visualization and assessment of the surface structure of the tracheobronchial system and the surrounding mediastinal structures and lesions. CONCLUSIONS: Hybrid rendering relieve the morphological assessment of anatomical and pathological changes without the need for time-consuming detailed analysis and presentation of source images. Performing virtual bronchoscopy with a transparent shaded-surface model offers a promising alternative to flexible fiberoptic bronchoscopy.

Hybrid 3D visualization of the chest and virtual endoscopy of the tracheobronchial system: possibilities and limitations of clinical application.

OBJECTIVE: A hybrid rendering method which combines a color-coded surface rendering method and a volume rendering method is described, which enables virtual endoscopic examinations using different representation models. MATERIALS AND METHODS: 14 patients with malignancies of the lung and mediastinum (n=11) and lung transplantation (n=3) underwent thin-section spiral computed tomography. The tracheobronchial system and anatomical and pathological features of the chest were segmented using an interactive threshold interval volume-growing segmentation algorithm and visualized with a color-coded surface rendering method. The structures of interest were then superimposed on a volume rendering of the other thoracic structures. For the virtual endoscopy of the tracheobronchial system, a shaded-surface model without color coding, a transparent color-coded shaded-surface model and a triangle-surface model were tested and compared. RESULTS: The hybrid rendering technique exploit the advantages of both rendering methods, provides an excellent overview of the tracheobronchial system and allows a clear depiction of the complex spatial relationships of anatomical and pathological features. Virtual bronchoscopy with a transparent color-coded shaded-surface model allows both a simultaneous visualization of an airway, an airway lesion and mediastinal structures and a quantitative assessment of the spatial relationship between these structures, thus improving confidence in the diagnosis of endotracheal and endobronchial diseases. CONCLUSIONS: Hybrid rendering and virtual endoscopy obviate the need for time consuming detailed analysis and presentation of axial source images. Virtual bronchoscopy with a transparent color-coded shaded-surface model offers a practical alternative to fiberoptic bronchoscopy and is particularly promising for patients in whom fiberoptic bronchoscopy is not feasible, contraindicated or refused. Furthermore, it can be used as a complementary procedure to fiberoptic bronchoscopy in evaluating airway stenosis and guiding bronchoscopic biopsy, surgical intervention and palliative therapy and is likely to be increasingly accepted as a screening method for people with suspected endobronchial malignancy and as control examination in the aftercare of patients with malignant diseases.

MR imaging of the wrist in rheumatoid arthritis using gadobenate dimeglumine.

OBJECTIVE: To determine the dosage of gadobenate dimeglumine (Gd-BOPTA) necessary for MRI of rheumatoid arthritis of the wrist. DESIGN AND PATIENTS: Seven wrists inflamed with rheumatoid arthritis were imaged using a dedicated 0.2-T MR unit. Four cumulative dosages of 0.0125, 0.025, 0.05 and 0.1 mmol/kg body weight (BW) Gd-BOPTA were tested. Three-dimensional T1-weighted gradient-recalled echo sequences (GRE; TR: 100 ms, TE: 18 ms, flip angle 90 degrees , 4:55 min) were acquired prior to an intravenous injection and after each additional dosage of Gd-BOPTA. Relative enhancement, signal-difference-to-noise ratios (SDNRs) and the size of the inflamed tissue were quantified. Three radiologists independently evaluated the image quality, the size and the contrast of the enhancing tissue. RESULTS: The readers agreed on a dose of 0.05 mmol/kg BW as satisfactory for the evaluation of the size of the inflammatory tissue and for determination of bone involvement (kappa = 0.9, P < 0.001). Highly inflammatory pannus was depicted with adequate image contrast using 0.025 mmol/kg BW Gd-BOPTA. According to the SDNR and relative enhancement findings, a dose of 0.05 mmol/kg BW suffices for both off-center and centered regions of tissue inflammation (t-test, P < 0.05). CONCLUSION: Gadolinium-BOPTA is an alternative contrast agent for MRI of rheumatoid disease. This study shows that a dose of 0.05 mmol/kg BW suffices at low field strength.

Correction of anterior thoracic wall deformities: improved planning by means of 3D-spiral-computed tomography.

Surgical correction of anterior thoracic wall deformities is an extensive procedure with considerable operative trauma. The procedure can be markedly supported by means of preoperative 3D spiral-computed tomography for the purpose of improved preoperative planning in order to avoid unnecessary tissue mobilization. It is also helpful to enhance the patients' comprehension regarding the operative requirements.

Hybrid rendering of the chest and virtual bronchoscopy [corrected].

Thin-section spiral computed tomography was used to acquire the volume data sets of the thorax. The tracheobronchial system and pathological changes of the chest were visualized using a color-coded surface rendering method. The structures of interest were then superimposed on a volume rendering of the other thoracic structures, thus producing a hybrid rendering. The hybrid rendering technique exploit the advantages of both rendering methods and enable virtual bronchoscopic examinations using different representation models. Virtual bronchoscopic examinations with a transparent color-coded shaded-surface model enables the simultaneous visualization of both the airways and the adjacent structures behind of the tracheobronchial wall and therefore, offers a practical alternative to fiberoptic bronchoscopy. Hybrid rendering and virtual endoscopy obviate the need for time consuming detailed analysis and presentation of axial source images.

Differentiation of malignant from benign solitary pulmonary lesions using chest radiography, spiral CT and HRCT.

OBJECTIVE: The aim of this prospective study was to summarize all of the qualitative and quantitative imaging criteria for the differentiation of solitary pulmonary lesions (SPLs) as malignant (MSPLs) or benign (BSPLs) described in the literature and to critically analyze the different characteristics in order to evaluate their clinical importance and usefulness as criteria for a discrimination during the primary diagnostic assessment of SPLs using chest radiography, spiral computed tomography (SCT) and high-resolution computed tomography (HRCT). MATERIALS AND METHODS: SPLs were examined, evaluated and then completely removed by surgery in 104 consecutive patients (MSPLs n=81, BSPLs n=23). No SPL was excluded by size. Chest radiography was performed with frontal and lateral views, SCT was carried out with a slice thickness of 8 mm and HRCT with a slice thickness of 1 mm and a 12-cm field of view. RESULTS: All the characteristics which enabled a reliable differentiation of MSPLs from BSPLs were characteristics which were observed significantly more frequently in MSPLs than BSPLs. Useful characteristics for the differentiation of MSPLs from BSPLs (1) using chest radiography were the indistinct edge (P<0.0001) and a ground-glass opacity of the lung parenchyma adjacent to the SPL (P<0. 05); (2) using SCT the presence of spicules (P<0.0005), the vessel sign (P<0.0005), necrotic areas (P<0.001), spicules extending to the visceral pleura (P<0.005), circumscribed pleural thickening (P<0. 005), inhomogeneity (P<0.01), a ground-glass opacity of the lung parenchyma adjacent to the SPL (P<0.01), the lesion density (P<0.05), pleural retraction (P<0.05) and the bronchus sign (P<0.05); and (3) using HRCT the presence of spicules (P<0.00005), spicules extending to the visceral pleura (P<0.0005), the vessel sign (P<0.0005), pleural retraction (P<0.001), circumscribed pleural thickening (P<0. 001), the bronchus sign (P<0.005), a ground-glass opacity of the lung parenchyma adjacent to the SPL (P<0.01), the lesion density (P<0.05) and the length of spicules (P<0.05). Using any one of the characteristics with a significance level of P<0.01, the identification of MSPLs (1) using chest radiography showed a sensitivity of 64.2% and a specificity of 82.6% (accuracy of 68.3%); (2) using SCT a sensitivity of 88.9% and a specificity of 60.9% (accuracy of 82.7%); and (3) using HRCT a sensitivity of 91.4% and a specificity of 56.5% (accuracy of 83.7%). CONCLUSIONS: Using chest radiography, SCT and HRCT, a precise morphological assessment of the periphery of the pulmonary lesion and the adjacent visceral pleura is necessary to distinguish MSPLs from BSPLs. In this respect SCT and HRCT are useful in differentiation of MSPLs from BSPLs. However, metastases strongly resembled benign lesions in terms of size and edge type and chronic inflammatory pseudotumors as a group mimic MSPLs.

[Hybrid 3D visualization and virtual endoscopy in cochlear implants]. / Hybride 3D-Visualisierung und virtuelle Endoskopie von Cochlea-Implantaten.

PURPOSE: The aim of this study was to demonstrate on a complex anatomical structure the possibilities and the advantages of a superimposition of a color-coded surface and volume rendering (hybrid rendering) method with the possibility of the performance of a virtual endoscopy. MATERIAL AND METHOD: In 6 patients with cochlear implants a high-resolution spiral computed tomography of the petrous bone was performed. The cochlear implants, the middle and inner ear structures were visualized using a color-coded surface rendering method, either shaded or as a grid. The petrous bone was visualized using a transparent volume rendering method. RESULTS: The hybrid 3D visualization uses the advantages of both the color-coded 3D surface and volume rendering method. In comparison to the axial source images, the hybrid 3D visualization thus facilitates a clearer representation and better assessment of the complex topographical relationship without loss of diagnostic information. The virtual endoscopy facilitates an intraluminal visualization and inspection of all color-coded 3D surface- and volume rendered structures. CONCLUSIONS: The hybrid rendering and virtual endoscopy make the morphological assessment of cochlear implants easier by the simultaneous visualization of the surrounding structures and thereby support the diagnostic imaging methods. This image processing method can be used pre-operatively for the individual planning, simulation, training and further development of surgical procedures and interventions and post-operatively for the control of the position and further developments of implants.

Evaluation of the middle and inner ear structures: comparison of hybrid rendering, virtual endoscopy and axial 2D source images.

Recent developments in 3D reconstructions can enhance the quality and diagnostic value of axial 2D image data sets with direct benefits for clinical practice. To show the possible advantages of a hybrid rendering method [color-coded 3D shaded-surface display (SSD)- and volume rendering method] with the possibility of virtual endoscopy we have specifically highlighted the use in relation to the middle and inner ear structures. We examined 12 patients with both normal findings and postoperative changes, using image data sets from high-resolution spiral computed tomography (HRSCT). The middle and inner ear was segmented using an interactive threshold interval density volume-growing method and visualized with a color-coded SSD rendering method. The temporal bone was visualized using a transparent volume rendering method. The 3D- and virtual reconstructions were compared with the axial 2D source images. The evaluated middle and inner ear structures could be seen in their complete form and correct topographical relationship, and the 3D- and virtual reconstructions indicated an improved representation and spatial orientation of these structures. A hybrid and virtual endoscopic method could add information and improve the value of imaging in the diagnosis and management of patients with middle or inner ear diseases making the understanding and interpretation of axial 2D CT image data sets easier. The introduction of an improved rendering algorithm aids radiological diagnostics, medical education, surgical planning, surgical training, and postoperative assessment.

An evaluation of the tumour markers, carcinoembryonic antigen (CEA), cytokeratin marker (CYFRA 21-1) and neuron-specific enolase (NSE) in the differentiation of malignant from benign solitary pulmonary lesions.

PURPOSE: The aim of this prospective study was to assess the diagnostic value of the tumour markers carcinoembryonic antigen (CEA), cytokeratin 19 fragment marker (CYFRA 21-1) and neuron-specific enolase (NSE) in the differentiation of malignant (MSPLs) from benign solitary pulmonary lesions (BSPLs). METHODS: Solitary pulmonary lesions (SPLs) were diagnosed using plain radiography and spiral computed tomography (SCT) and then completely removed by surgery in 104 consecutive patients (MSPLs; n = 81, BSPLs; n = 23). The serum concentrations of the tumour markers were determined 1-3 days prior to surgery by ELISA for CEA and CYFRA 21-1 and by IRMA for NSE using commercially available assay kits. The cut-off values were set at 3 ng/ml (for non-smokers) and 5 ng/ml (for smokers) for CEA, at 3.3 ng/ml for CYFRA 21-1 and at 12.5 ng/ml for NSE. RESULTS: MSPLs were identified with a sensitivity between 13.6 and 45.7%, a specificity between 87.0 and 100% and an accuracy between 32.7 and 54.8%. Using the tumour markers alone, the highest sensitivity (27.2%) and accuracy (40.4%) was found with CEA, the highest specificity (100%) with CYFRA 21-1 and with NSE. Primary lung cancers (n = 39) were identified with a sensitivity between 17.9 and 61.5%, a specificity between 87.0 and 100% and an accuracy between 48.4 and 71.0%. Using the tumour markers alone, the highest sensitivity (35.9%) and accuracy (59.7%) was found with CYFRA 21-1, the highest specificity (100%) with CYFRA 21-1 and with NSE. The combination of all three tumour markers resulted in a greater sensitivity and greater diagnostic accuracy but a loss in specificity compared with CYFRA 21-1 and NSE. CONCLUSION: The use of the tumour markers alone or in combination showed a low sensitivity and low accuracy for the diagnostic differentiation of MSPLs from BSPLs and primary lung cancers from BSPLs. However, both CYFRA 21-1 and NSE exhibited a specificity of 100% and may be useful complements to standard clinical imaging methods.