Retinal template matching and registration is an important challenge in teleophthalmology with low-cost imaging devices. However, the images from such devices generally have a small field of view (FOV) and image quality degradations, making matching difficult. In this paper, we develop an efficient and accurate retinal matching technique that combines dimension reduction and mutual information (MI), called RetinaMatch. The dimension reduction initializes the MI optimization as a coarse localization process, which narrows the optimization domain and avoids local optima. The effectiveness of RetinaMatch is demonstrated on the open fundus image database STARE with simulated reduced FOV and anticipated degradations, and on retinal images acquired by adapter-based optics attached to a smartphone. RetinaMatch achieves a success rate over 94% on human retinal images with the matched target registration errors below 2 pixels on average, excluding the observer variability, outperforming standard template matching solutions. In the application of measuring vessel diameter repeatedly, single pixel errors are expected. In addition, our method can be used in the process of image mosaicking with area-based registration, providing a robust approach when feature-based methods fail. To the best of our knowledge, this is the first template matching algorithm for retina images with small template images from unconstrained retinal areas. In the context of the emerging mixed reality market, we envision automated retinal image matching and registration methods as transformative for advanced teleophthalmology and long-term retinal monitoring.
Esophageal adenocarcinoma (EAC) is a deadly cancer that is rising rapidly in incidence. The early detection of EAC with curative intervention greatly improves the prognoses of patients. A scanning fiber endoscope (SFE) using fluorescence-labeled peptides that bind rapidly to epidermal growth factor receptors showed a promising performance for early EAC detection. Target-to-background (T/B) ratios were calculated to quantify the fluorescence images for neoplasia lesion classification. This T/B calculation is generally based on lesion segmentation with the Chan–Vese algorithm, which may require hyperparameter adjustment when segmenting frames with different brightness and contrasts, which impedes automation to real-time video. Deep learning models are more robust to these changes, while accurate pixel-level segmentation ground truth is challenging to establish in the medical field. Since within our dataset the ground truth contained only a frame-level diagnosis, we proposed a computer-aided diagnosis (CAD) system to calculate the T/B ratio in real time. A two-step process using convolutional neural networks (CNNs) was developed to achieve automatic suspicious frame selection and lesion segmentation for T/B calculation. In the segmentation model training for Step 2, the lesion labels were generated with a manually tuned Chan–Vese algorithm using the labeled and predicted suspicious frames from Step 1. In Step 1, we designed and trained deep CNNs to select suspicious frames using a diverse and representative set of 3427 SFE images collected from 25 patient videos from two clinical trials. We tested the models on 1039 images from 10 different SFE patient videos and achieved a sensitivity of 96.4%, a specificity of 96.6%, a precision of 95.5%, and an area under the receiver operating characteristic curve of 0.989. In Step 2, 1006 frames containing suspicious lesions were used for training for fluorescence target segmentation. The segmentation models were tested on two clinical datasets with 100 SFE frames each and achieved mean intersection-over-union values of 0.89 and 0.88, respectively. The T/B ratio calculations based on our segmentation results were similar to the manually tuned Chan–Vese algorithm, which were 1.71 ± 0.22 and 1.72 ± 0.28, respectively, with a p-value of 0.872. With the graphic processing unit (GPU), the proposed two-step CAD system achieved 50 fps for frame selection and 15 fps for segmentation and T/B calculation, which showed that the frame rejection in Step 1 improved the diagnostic efficiency. This CAD system with T/B ratio as the real-time indicator is designed to guide biopsies and surgeries and to serve as a reliable second observer to localize and outline suspicious lesions highlighted by fluorescence probes topically applied in organs where cancer originates in the epithelia.
Purpose: Handling low-quality and few-feature medical images is a challenging task in automatic panorama mosaicking. Current mosaicking methods for disordered input images are based on feature point matching, whereas in this case intensity-based registration achieves better performance than feature-point registration methods. We propose a mosaicking method that enables the use of mutual information (MI) registration for mosaicking randomly ordered input images with insufficient features. Approach: Dimensionality reduction is used to map disordered input images into a low dimensional space. Based on the low dimensional representation, the image global correspondence can be recognized efficiently. For adjacent image pairs, we optimize the MI metric for registration. The panorama is then created after image blending. We demonstrate our method on relatively lower-cost handheld devices that acquire images from the retina in vivo, kidney ex vivo, and bladder phantom, all of which contain sparse features. Results: Our method is compared with three baselines: AutoStitch, “dimension reduction + SIFT,” and “MI-Only.” Our method compared to the first two feature-point based methods exhibits 1.25 (ex vivo microscope dataset) to two times (in vivo retina dataset) rate of mosaic completion, and MI-Only has the lowest complete rate among three datasets. When comparing the subsequent complete mosaics, our target registration errors can be 2.2 and 3.8 times reduced when using the microscopy and bladder phantom datasets. Conclusions: Using dimensional reduction increases the success rate of detecting adjacent images, which makes MI-based registration feasible and narrows the search range of MI optimization. To the best of our knowledge, this is the first mosaicking method that allows automatic stitching of disordered images with intensity-based alignment, which provides more robust and accurate results when there are insufficient features for classic mosaicking methods.
14 May 2022 in Journal of Medical Robotics Research
Telecystoscopy can lower the barrier to access critical urologic diagnostics for patients around the world. A major challenge for robotic control of flexible cystoscopes and intuitive teleoperation is the pose estimation of the scope tip. We propose a novel real-time camera localization method using video recordings from a prior cystoscopy and 3D bladder reconstruction to estimate cystoscope pose within the bladder during follow-up telecystoscopy. We map prior video frames into a low-dimensional space as a dictionary so that a new image can be likewise mapped to efficiently retrieve its nearest neighbor among the dictionary images. The cystoscope pose is then estimated by the correspondence among the new image, its nearest dictionary image, and the prior model from 3D reconstruction. We demonstrate performance of our methods using bladder phantoms with varying fidelity and a servo-controlled cystoscope to simulate the use case of bladder surveillance through telecystoscopy. The servo-controlled cystoscope with 3 degrees of freedom (angulation, roll, and insertion axes) was developed for collecting cystoscope videos from bladder phantoms. Cystoscope videos were acquired in a 2.5D bladder phantom (bladder-shape cross-section plus height) with a panorama of a urothelium attached to the inner surface. Scans of the 2.5D phantom were performed in separate arc trajectories each of which is generated by actuation on the angulation with a fixed roll and insertion length. We further included variance in moving speed, imaging distance and existence of bladder tumors. Cystoscope videos were also acquired in a water-filled 3D silicone bladder phantom with hand-painted vasculature. Scans of the 3D phantom were performed in separate circle trajectories each of which is generated by actuation on the roll axis under a fixed angulation and insertion length. These videos were used to create 3D reconstructions, dictionary sets, and test data sets for evaluating the computational efficiency and accuracy of our proposed method in comparison with a method based on global Scale-Invariant Feature Transform (SIFT) features, named SIFT-only. Our method can retrieve the nearest dictionary image for 94–100% of test frames in under 55 ms per image, whereas the SIFT-only method can only find the image match for 56–100% of test frames in 6000–40000 ms per image depending on size of the dictionary set and richness of SIFT features in the images. Our method, with a speed of around 20 Hz for the retrieval stage, is a promising tool for real-time image-based scope localization in robotic cystoscopy when prior cystoscopy images are available.