Image Fusion Procedure Research Articles

Multi-exposure fused light field image quality assessment for dynamic scenes: Benchmark dataset and objective metric

The luminance dynamic range in natural scenes is exceedingly wide. However, capturing the wide dynamic range information of natural scenes through a single exposure is challenging for commercial light field cameras. One solution to expand the luminance dynamic range of captured light field images is to employ multi-exposure imaging techniques. Nevertheless, this approach may introduce various distortions during the image fusion procedure, especially for dynamic scenes. To evaluate the visual quality of multi-exposure fused light field images (MEFLFIs) and compare the performance of different multi-exposure light field fusion methods, we initially conduct a subjective quality assessment on MEFLFIs for dynamic scenes and establish the first MEFLFI benchmark dataset, which comprises 480 fused light field images generated by eleven state-of-the-art multi-exposure fusion algorithms and two tone-mapping algorithms, along with their corresponding subjective scores. Subsequently, a novel specially designed objective quality assessment metric for MEFLFIs is developed, which incorporates both local–global joint features and angular quality-aware features. The key contribution of this work lies in establishing a pioneering benchmark dataset for MEFLFIs along with a novel tailored objective quality metric. Experimental results conducted over our established MEFLFI database demonstrate that the proposed metric achieves superior performance, with Pearson linear correlation coefficient (PLCC) and Spearman rank correlation coefficient (SROCC) values of 0.8895 and 0.8658 respectively, representing a significant improvement of approximately 5.7 % and 4.9 % compared to the second-best specifically-designed LFI quality metric available. Furthermore, ablation experiments have confirmed that the integration of local–global joint features and angular features significantly enhances the performance of our objective metric, resulting in a better alignment with human visual perception.

Fourier phase analysis combined with a fusion scheme in long pulse thermography

Fourier phase analysis (FPA) is a common routine in long pulse thermography (LPT). However, the change of appearance in maximum phase difference in the Fourier domain caused by different depths of defects results in low contrast images with fuzzy defect edges, limiting its application majorly. In this work, the detection capability of FPA is studied by modelling the image sequence length as a function of the defect depth, thereby quantifying the expected image sequence length for inspecting the defects at a specific depth and then, a strategy for selecting the optimal image sequence length is proposed by establishing the relationship between the maximum phase difference and the image sequence length. Followed by an image fusion procedure, the phase difference map with high-contrast defect edges can be obtained from the optimal image sequence. In addition, we show that the precooling treatment before detection is an effective way to improve the image quality for LPT. Experiments have been conducted in identifying defects in a typical glass fiber reinforced polymer (GFPR) with size of 200 × 150 × 6 mm3, where continuous thermal excitation for 100 s was provided by a pair of 1 kW halogen lamps and the images were recorded by an infrared camera with wavelength band of 8–14 μm. The resulting phase difference maps clearly present the edges of all defects with a signal-to-noise ratio greater than 43 dB, demonstrating the effectiveness of the proposed method in improving the image quality in LPT.

A Feature Extraction Using Probabilistic Neural Network and BTFSC-Net Model with Deep Learning for Brain Tumor Classification

Brain Tumor Fusion-based Segments and Classification-Non-enhancing tumor (BTFSC-Net) is a hybrid system for classifying brain tumors that combine medical image fusion, segmentation, feature extraction, and classification procedures. to reduce noise from medical images, the hybrid probabilistic wiener filter (HPWF) is first applied as a preprocessing step. Then, to combine robust edge analysis (REA) properties in magnetic resonance imaging (MRI) and computed tomography (CT) medical images, a fusion network based on deep learning convolutional neural networks (DLCNN) is developed. Here, the brain images' slopes and borders are detected using REA. To separate the sick region from the color image, adaptive fuzzy c-means integrated k-means (HFCMIK) clustering is then implemented. To extract hybrid features from the fused image, low-level features based on the redundant discrete wavelet transform (RDWT), empirical color features, and texture characteristics based on the gray-level cooccurrence matrix (GLCM) are also used. Finally, to distinguish between benign and malignant tumors, a deep learning probabilistic neural network (DLPNN) is deployed. according to the findings, the suggested BTFSC-Net model performed better than more traditional preprocessing, fusion, segmentation, and classification techniques. Additionally, 99.21% segmentation accuracy and 99.46% classification accuracy were reached using the proposed BTFSC-Net model. earlier approaches have not performed as well as our presented method for image fusion, segmentation, feature extraction, classification operations, and brain tumor classification. These results illustrate that the designed approach performed more effectively in terms of enhanced quantitative evaluation with better accuracy as well as visual performance.

Study of image fusion optimization techniques for medical applications

Image fusion is becoming increasingly important in computer vision activities because of the larger number of capture methods. The integration of different views like multi view, multi temporal view and information which is larger together termed as image fusion (IF). In image fusion the information of multiple-sensors are converted into a unified image while maintaining the fidelity of critical characteristics.Healthcare image fusion procedures are used to improve picture quality by achieving the conspicuous characteristics in the fusion findings. As a result, they increase the practical usefulness of medical pictures for issue assessment and identification. Fusion of medical images are generally evaluated using the modalities like MRI-Magnetic Resonance Imaging, MRA-Magnetic Resonance Angiogram, PET-Positron Emission Tomography, SPET-Structural Positron Emission Tomography,CT-Computed Tomography, and SPECT-Single-Photon Emission Computed Tomography. Neural network and optimization techniques help in improving the quality of fused image. In this paper various fusion techniques are studied. Along with different fuson approaches, the outcomes of diverse research projects are contrasted in terms of how the pulse coupled neural network is employed and different optimization techniques. The Pulse Couple Neural Networks (PCNN) using various optimization techniques are compared. Among which the swarming mechanism of salps improves the performance of the system. The PCNN is combined with SSO algorithm and evaluated the results.From the results it is shown that PCNN- Salp Swarm Optimization (SSO)ishaving good value of Peak Signal to Noise Ratio (PSNR) with 45.93 and Structural Similarity Index(SSIM) is 0.996.

Multimodal Classification on PET/CT Image Fusion for Lung Cancer: A Comprehensive Survey

Medical image fusion has become essential for accurate diagnosis. For example, a lung cancer diagnosis is currently conducted with the help of multimodality image fusion to find anatomical and functional information about the tumor and metabolic measurements to identify the lung cancer stage and metastatic information of the disease. Generally, the success of multimodality imaging for lung cancer diagnosis is due to the combination of PET and CT imaging advantages while minimizing their respective limitations. However, medical image fusion involves the registration of two different modalities, which is time-consuming and technically challenging, and it is a cause of concern in a clinical setting. Therefore, the paper's main objective is to identify the most efficient medical image fusion techniques and the recent advances by conducting a collective survey. In addition, the study delves into the impact of deep learning techniques for image fusion and their effectiveness in automating the image fusion procedure with better image quality while preserving essential clinical information.

Detail-Injection-Model-Inspired Deep Fusion Network for Pansharpening

Pansharpening is an image fusion procedure, which aims to produce a high spatial resolution multispectral image by combining a low spatial resolution multispectral image and a high spatial resolution panchromatic image. The most popular and successful paradigm for pansharpening is the framework known as detail injection, while it cannot fully exploit complex and non-linear complementary features of both images. In this paper, we propose a detail injection model inspired deep fusion network for pansharpening (DIM-FuNet). Firstly, by treating pansharpening as a complicated and non-linear details learning and injection problem, we establish a unified optimizing detail-injection model with triple detail fidelity terms: 1) a band-dependent spatial detail fidelity term, 2) a local detail fidelity term and 3) a complicated details synthesis term. Secondly, the model is optimized via the iterative gradient descent and unfolded into a deep convolutional neural network. Subsequently, the unrolling network has triple branches, in which, a point-wise convolutional sub-network, a depth-wise convolutional sub-network are corresponding to the former two detail constrained terms, and an adaptive weighted reconstruction module with a fusion sub-network to aggregate details of two branches and synthesis the final complicated details. Finally, the deep unrolling network is trained in end-to-end manners. Different from traditional deep fusion networks, the architecture design of DIM-FuNet is guided by the optimizing model and thus promotes better interpretability. Experimental results on reduced and full-resolution demonstrate the effectiveness of the proposed DIM-FuNet which achieves the best performance compared with the state-of-the-art pansharpening method.

Analysis of Discrete Wavelet Transforms Variants for the Fusion of CT and MRI Images

Background: Medical image fusion methods are applied to a wide assortment of medical fields, for example, computer-assisted diagnosis, telemedicine, radiation treatment, preoperative planning, and so forth. Computed Tomography (CT) is utilized to scan the bone structure, while Magnetic Resonance Imaging (MRI) is utilized to examine the soft tissues of the cerebrum. The fusion of the images obtained from the two modalities helps radiologists diagnose the abnormalities in the brain and localize the position of the abnormality concerning the bone. Methods: Multimodal medical image fusion procedure contributes to the decrease of information vulnerability and improves the clinical diagnosis exactness. The motive is to protect salient features from multiple source images to produce an upgraded fused image. The CT-MRI image fusion study made it conceivable to analyze the two modalities straightforwardly. Several states of the art techniques are available for the fusion of CT & MRI images. The discrete wavelet transform (DWT) is one of the widely used transformation techniques for the fusion of images. However, the efficacy of utilization of the variants of wavelet filters for the decomposition of the images, which may improve the image fusion quality, has not been studied in detail. Therefore the objective of this study is to assess the utility of wavelet families for the fusion of CT and MRI images. In this paper investigation on the efficacy of 8 wavelet families (120 family members) on the visual quality of the fused CT & MRI image has been performed. Further, to strengthen the quality of the fused image, two quantitative performance evaluation parameters, namely classical and gradient information, have been calculated. Results: Experimental results demonstrate that amongst the 120 wavelet family members (8 wavelet families), db1, rbio1.1, and Haar wavelets have outperformed other wavelet family members in both qualitative and quantitative analysis. Conclusion: Quantitative and qualitative analysis shows that the fused image may help radiologists diagnose the abnormalities in the brain and localize the position of the abnormality concerning the bone more easily. For further improvement in the fused results, methods based on deep learning may be tested in the future.

Hyperspectral Image Superresolution via Structure-Tensor-Based Image Matting

Hyperspectral (HS) imaging has achieved breakthroughs in many applications, such as remote sensing and object recognition. However, the spatial resolution of HS images is still insufficient due to the limitations of sensor technology and cost. In this article, we propose an HS image superresolution method that combines low-resolution (LR) HS images and high-resolution (HR) panchromatic (PAN) images. To exploit the spectral signatures in the LR-HS images while introducing details from the HR-PAN images during the image fusion procedure, an image matting model is used to fuse the original LR-HS images and the HR-PAN images. Specifically, to preserve the spectral components during the fusion procedure, two different alpha channels in the image matting model are generated based on the HS and PAN image structure tensors, which suppress spectral distortion and improve the quality of the reconstructed HR-HS image. Experimental results based on public datasets demonstrate the advantage of our proposed method in both preserving spectral information and enhancing HS image spatial resolution.

Metrological analysis of the three-dimensional reconstruction based on close-range photogrammetry and the fusion of long-wave infrared and visible-light images

This work proposes evaluating statistically the metrological performance of three-dimensional reconstructions built with fused long-wavelength infrared (LWIR) and visible-light (VL) images. The image fusion procedure was essentially based on two-dimensional wavelet transform and two pixel-level fusion rules: the maximum intensity level, presented in a previous work of the authors, and a new fusion rule, which replaces the VL information with the LWIR information in the region of the measured object on the images. The reconstructions of a translucent cube were performed with a point triangulation-based procedure and its dimension measurements were employed as evaluation criteria. The results show that the fused images have more contrast but also more artifacts. The fusion procedures generated denser reconstructions with at least 34.83% more points. Considering the metrological result, reconstructions with only visible-light images resulted in maximal 89.31% less measurement bias but at least 47.25% more uncertainty than the fusion ones. The new fusion rule provided the best results, with more points in the dense cloud and lower uncertainty. The work is important to provide a metrologically viable alternative for three-dimensional reconstruction of objects in situations of low contrast or poor texture information in the visible spectrum, and in which no target can be applied to the inspected part.

Image Fusion Metrics – A Systematic Study

When two input images with different focus areas are combined in such a way that resultant image is leveraged with all the required and pertinent information to fulfil the purpose of study and research as well as per the observer, then all this process is known as multi-focus image fusion procedure. But how will be able to know that fused image is as per the required standard? To fulfil this objective, having performance metrics in the field of image fusion. To assess the quality of the resultant image, performance metrics are used as an evaluation parameter estimators. After completing fusion process, the resultant fused image is whether as per the requirement or not, whether it possesses all the pertinent information, to make sure there is no loss of information in the final resultant image, the need of fusion performance metrics comes into the picture. In this paper, subjective as well as objective fusion metrics have been studied to have an in-depth understanding of the suitability of the respective performance parameters so that it can be easily concluded that at what stage which metrics will provide the better results.

PC SWT A Hybrid Image Fusion Algorithm of Stationary Wavelet Transform and Principal Component Analysis

Image fusion is viewed as perhaps the best procedure to confine the level of uncertainty and convey a significant feeling of picture lucidity. It is a strategy of combining the appropriate information/data from a group of pictures into a solitary resultant (intertwined) picture that would render higher picture proficiency and clarity. Until now, the image fusion procedures looked like Discrete Wavelet Transform (DWT) or pixel-based methodologies. These already established methods have limited effectiveness. Also, they fail to deliver the typical outcomes like edge perseverance, spatial resolution, and shift-invariance. To get rid of these demerits, in this paper, we have proposed a hybrid approach called Principal Component Stationary Wavelet Transform (PC-SWT) that combines Principal Component Analysis (PCA) and Stationary Wavelet Transform. SWT is an algorithm that defines the wavelet transformation to compensate for the absence of translation invariance in DWT. PCA is a methodical approach that utilizes an orthogonal transformation in order to transform a group of perceptions of possibly correlated values into the principal components, which are linearly uncorrelated variables. When compared to conventional methods, PC-SWT intends to obtain a more efficient, clear, and superior quality image. This fused image is expected to have all of its preserved edges as well as its spatial resolution. In addition to this, it can also be used to deal with shift-invariance.

Medical image fusion based on laws of texture energy measures in stationary wavelet transform domain

AbstractMedical image fusion is widely used in various clinical procedures for the precise diagnosis of a disease. Image fusion procedures are used to assist real‐time image‐guided surgery. These procedures demand more accuracy and less computational complexity in modern diagnostics. Through the present work, we proposed a novel image fusion method based on stationary wavelet transform (SWT) and texture energy measures (TEMs) to address poor contrast and high‐computational complexity issues of fusion outcomes. SWT extracts approximate and detail information of source images. TEMs have the capability to capture various features of the image. These are considered for fusion of approximate information. In addition, the morphological operations are used to refine the fusion process. Datasets consisting of images of seven patients suffering from neurological disorders are used in this study. Quantitative comparison of fusion results with visual information fidelity‐based image fusion quality metric, ratio of spatial frequency error, edge information‐based image fusion quality metric, and structural similarity index‐based image fusion quality metrics proved the superiority. Also, the proposed method is superior in terms of average execution time to state‐of‐the‐art image fusion methods. The proposed work can be extended for fusion of other imaging modalities like fusion of functional image with an anatomical image. Suitability of the fused images by the proposed method for image analysis tasks needs to be studied.

Medical Image Fusion using Transform Based Fusion Techniques

The principal resolution of the image fusion is to merging indication from different images; CT (Computed Tomography) scan and an MRI (Magnetic Resonance Imaging) and to obtain more informative image. In this paper various transform based fusion methods like; discrete wavelet transform (DWT) and two specialisms of discrete cosine transform (DCT); DCT variance and DCT variance with consistency verification (DCT variance with CV) and stationary wavelet transform (SWT) image fusion procedures are instigated and associated in terms of image evidence. Fused outcomes attained from these fusion techniques are evaluated through distinctive evaluation metrics. A fused result accomplished from DCT variance with CV followed by DCT variance out performs DWT and SWT based image fusion methodologies. The potentiality of DCT features creates value-added evidence in the output fused image trailed by fused results proficient from DWT and SWT based image fusion methods. The discrete cosine transforms (DCT) stranded methods of image fusion are auxiliary accurate and concert leaning in real time solicitations by energy forte of DCT originated ideologies of stationary images. In this effort, a glowing systematic practice for fusion of multi-focus images based on DCT and its flavors are obtainable and demonstrated that DCT grounded fused outcomes exceed other fusion methodologies.

Radiotherapy dose verification on a customised head and neck perspex phantom

IMRT dose planned for head and neck radiotherapy was verified using a customised acrylic head-and-neck phantom. The dosimeters used were calibrated Gafchromic EBT2 film and metal–oxide–semiconductor-field-effect-transistor (MOSFET). Target volumes (TV) and organs-at-risk (OAR) which were previously contoured by an oncologist on selected nasopharynx (NPC) patients were transferred to this phantom by an image fusion procedure. Three radiotherapy plans were done: Plan1 with 7-fields intensity-modulated radiotherapy (IMRT) of prescribed dose 70 Gy using 33 fractions; Plan2 with 7-fields IMRT plan at 70 Gy and 35 fractions; and Plan3 which was a mid-plane-dose (MPD) plan of 66 Gy at 33 fractions. The dose maps were first verified using MapCheck2 by SNC-PatientTM software. The passing rates from gamma analysis were 97.7% (Plan1), 93.1% (Plan2) and 100% (Plan3). Percentage difference between Treatment Planning System (TPS) calculated dose and MOSFET measured dose was comparatively higher than those from EBT2. Calculated dose and EBT2 measured doses showed differences of within the range of ±3% for TV and <±10% for OARs. However MOSFET had differences of within the range of ±6% for TV and within the range of ±10% for OARs between measured and planned doses. An overdose treatment may occur as TPS calculated doses were lower than the measured doses in these plans. This may be due to the effects of leaf leakage, leaf scatter and photon backscatter into the measuring tools (Pawlicki et al., 1999 and Ma et al., 2000). More IMRT plans have to be studied to validate this conclusion. However, the dose measurements were still within the 10% tolerance (AAPM Task Group 119). In conclusion, both GafchromicEBT2 film and MOSFET are suitable for IMRT radiotherapy dosimetry.

MP16-06 DIAGNOSTIC ACCURACY OF MAGNETIC RESONANCE IMAGING (MRI) 5-POINT LIKERT SCORING SYSTEM EVALUATED BY THE RESULT OF MRI/ULTRASONOGRAPHY FUSION TARGETED BIOPSY OF THE PROSTATE

INTRODUCTION AND OBJECTIVES: The aim of this study was to evaluate the accuracy of magnetic resonance imaging (MRI) scoring rated by the different radiologists for prostate cancer detection using MRI/trans-rectal ultrasonography (US) fusion targeted biopsy as reference standard. METHODS: We retrospectively reviewed 762 patients who underwent 3-Tesla multi-parametric (mp)-MRI and the following MRI/US fusion biopsy, all of which were performed by experienced urologists (10/2012-8/2015). We excluded patients in whom MRI did not identify any suspicious lesions and radiologists who reported in less than 50 cases. Finally, 648 patients with 1255 suspicious lesions rated by 8 radiologists were included in this study. The mp-MRIs were reported on a 5-point Likert scale of suspicion. The UroStation (Koelis, France) was used for the image fusion procedure and histology of all lesions was obtained by MRI/US fusion targeted biopsy. Clinically significant cancer was defined as biopsy Gleason score 7. RESULTS: Median age was 64 years, pre-biopsy prostatespecific antigen (PSA) level was 6.93 ng/ml and estimated prostate volume was 52.1 ml. Of 1255 suspicious lesions on MRI, 83% (n1⁄41036) were rated as grade 1-3 (lower suspicion) and 17% (n1⁄4219) were grade 4-5 (higher suspicion). According to the 5-point Likert system of grade from 1 to 5, overall cancer detection rate were 12%, 13%, 22%, 52%, 59%, and significant cancer detection rate were 4%, 4%, 12%, 33%, 48%, respectively. Grading of a 5-point scale of suspicion on MRI showed strong positive correlation with overall cancer detection rate (R1⁄40.946, p1⁄40.05) and significant cancer detection rate (R1⁄40.945, p1⁄40.05). In comparison between grade 4-5 lesions versus grade 1-3 lesions, the targeted biopsies from grade 4-5 lesions had a higher overall cancer detection rate (55% vs 17%, p<0.001) and significant cancer detection rate (40% vs 7%, p<0.001), respectively. In comparison between the 4 more experienced radiologists for mpMRI-prostate versus the other 4 less experienced radiologists, statistical differences were noted in overall cancer detection rate (62% vs 43%, p1⁄40.007) and significant cancer detection rate (45% vs 31%, p1⁄40.038) in the grade 4-5 lesions. CONCLUSIONS: Overall cancer detection rate and significant cancer detection rate strongly correlated with 5-point scale of suspicion on MRI. Among the radiologists with different level of experience for mpMRI-prostate, there were significant differences in overall cancer detection rate and significant cancer detection rate.

Urgent endovascular repair for ruptured aortic aneurysm using computed tomography image fusion

Abdominal aortic aneurysm rupture remains a cardiovascular catastrophe with strikingly high morbidity and mortality rates. Endovascular aneurysm repair management has recently emerged as a valuable treatment modality for ruptured abdominal aortic aneurysm, but better outcomes have to be reached. Image fusion may potentially improve perioperative outcomes in selected patients, guiding navigation and device implantation and limiting contrast dosage during interventions. We report an 83-year-old man presenting with an 80-mm infrarenal aortic aneurysm rupture suitable for endovascular aneurysm repair. Endovascular navigation and stent graft deployment were achieved using computed tomography image fusion for the first reported case in English.

Assessment of image fusion procedures using entropy, image quality, and multispectral classification

The use of disparate data sources within a pixel level image fusion procedure has been well documented for pan-sharpening studies. The present paper explores various image fusion procedures for the fusion of multi-spectral ASTER data and a RadarSAT-1 SAR scene. The research sought to determine which fusion procedure merged the largest amount of SAR texture into the ASTER scenes, while also preserving the spectral content. An additional application based maximum likelihood classification assessment was also undertaken. Three SAR scenes were tested namely, one backscatter scene and two textural measures calculated using grey level co-occurrence matrices (GLCM). Each of these were fused to the ASTER data using the following established approaches; Brovey transformation, Intensity Hue and Saturation, Principal Component Substitution, Discrete wavelet transformation, and a modified discrete wavelet transformation using the IHS approach. Resulting data sets were assessed using qualitative and quantitative (entropy, universal image quality index, maximum likelihood classification) approaches. Results from the study indicated that while all post fusion data sets contained more information (entropy analysis), only the frequency-based fusion approaches managed to preserve the spectral quality of the original imagery. Furthermore results also indicated that the textural (mean, contrast) SAR scenes did not add any significant amount of information to the post-fusion imagery. Classification accuracy was not improved when comparing ASTER optical data and pseudo optical bands generated from the fusion analysis. Accuracies range from 68.4% for the ASTER data to well below 50% for the component substitution methods. Frequency based approaches also returned lower accuracies when compared to the unfused optical data. The present study essentially replicated (pan-sharpening) studies using the high resolution SAR scene as a pseudo panchromatic band.

An automated robot arm system for small animal tissue biopsy under dual-image modality

The ability to non-invasively monitor cell biology in vivo is one of the most important goals of molecular imaging. Imaging procedures could be inter-subject performed repeatedly at different investigating stages; thereby need not sacrifice small animals during the entire study period. Thus, the ultimate goal of this study was to design a stereotactic image-guided system for small animals and integrated it with an automatic robot arm for in vivo tissue biopsy analysis. The system was composed of three main parts, including one small animal stereotactic frame, one imaging-fusion software and an automatic robot arm system. The system has been thoroughly evaluated with three components; the robot position accuracy was 0.05±0.02 mm, the image registration accuracy was 0.37±0.18 mm and the system integration was satisfactorily within 1.20±0.39 mm of error. From these results, the system demonstrated sufficient accuracy to guide the micro-injector from the planned delivery routes into practice. The entire system accuracy was limited by the image fusion and orientation procedures, due to its nature of the blurred PET imaging obtained from the small objects. The primary improvement is to acquire as higher resolution as possible the fused imaging for localizing the targets in the future.

Use of single- and multi-source image fusion for statistical decision-making

In this paper, we explore the use of single- and multi-source image fusion for statistical decision-making. On the basis of usability and loss functions, two image fusion procedures are formulated. For each of these, an optimality criterion is defined. First, attention focuses on fusion of different bands within a single image, which is segmented and classified. Second, a geostatistical procedure is identified to fuse a D-InSAR image with point measurements on leveling. A decision of the optimal fusion procedure is done using a qualitative decision criterion.

Klinischer Einsatz der digitalen retrospektiven Bildfusion von CT, MRT, FDG-PET und SPECT - Anwendungsgebiete und Ergebnisse

To evaluate the feasibility and the clinical benefits of retrospective digital image fusion (PET, SPECT, CT and MRI). In a prospective study, a total of 273 image fusions were performed and evaluated. The underlying image acquisitions (CT, MRI, SPECT and PET) were performed in a way appropriate for the respective clinical question and anatomical region. Image fusion was executed with a software program developed during this study. The results of the image fusion procedure were evaluated in terms of technical feasibility, clinical objective, and therapeutic impact. The most frequent combinations of modalities were CT/PET (n = 156) and MRI/PET (n = 59), followed by MRI/SPECT (n = 28), CT/SPECT (n = 22) and CT/MRI (n = 8). The clinical questions included following regions (more than one region per case possible): neurocranium (n = 42), neck (n = 13), lung and mediastinum (n = 24), abdomen (n = 181), and pelvis (n = 65). In 92.6 % of all cases (n = 253), image fusion was technically successful. Image fusion was able to improve sensitivity and specificity of the single modality, or to add important diagnostic information. Image fusion was problematic in cases of different body positions between the two imaging modalities or different positions of mobile organs. In 37.9 % of the cases, image fusion added clinically relevant information compared to the single modality. For clinical questions concerning liver, pancreas, rectum, neck, or neurocranium, image fusion is a reliable method suitable for routine clinical application. Organ motion still limits its feasibility and routine use in other areas (e. g., thorax).