3 July 2020 in BMC Urology
The aims of this study were to determine the predictive value of decision support analysis for the shock wave lithotripsy (SWL) success rate and to analyze the data obtained from patients who underwent SWL to assess the factors influencing the outcome by using machine learning methods. We retrospectively reviewed the medical records of 358 patients who underwent SWL for urinary stone (kidney and upper-ureter stone) between 2015 and 2018 and evaluated the possible prognostic features, including patient population characteristics, urinary stone characteristics on a non-contrast, computed tomographic image. We performed 80% training set and 20% test set for the predictions of success and mainly used decision tree-based machine learning algorithms, such as random forest (RF), extreme gradient boosting trees (XGBoost), and light gradient boosting method (LightGBM). In machine learning analysis, the prediction accuracies for stone-free were 86.0, 87.5, and 87.9%, and those for one-session success were 78.0, 77.4, and 77.0% using RF, XGBoost, and LightGBM, respectively. In predictions for stone-free, LightGBM yielded the best accuracy and RF yielded the best one in those for one-session success among those methods. The sensitivity and specificity values for machine learning analytics are (0.74 to 0.78 and 0.92 to 0.93) for stone-free and (0.79 to 0.81 and 0.74 to 0.75) for one-session success, respectively. The area under curve (AUC) values for machine learning analytics are (0.84 to 0.85) for stone-free and (0.77 to 0.78) for one-session success and their 95% confidence intervals (CIs) are (0.730 to 0.933) and (0.673 to 0.866) in average of methods, respectively. We applied a selected machine learning analysis to predict the result after treatment of SWL for urinary stone. About 88% accurate machine learning based predictive model was evaluated. The importance of machine learning algorithm can give matched insights to domain knowledge on effective and influential factors for SWL success outcomes.
1 January 2021 in Korean Journal of Radiology
Park HS, et al. Korean J Radiol. 2020 Jul;21:e178. https://doi.org/10.3348/kjr.2020.0051
15 October 2019 in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization
Automatic annotation for three-dimensional (3D) cephalometric analysis has been limited by computational complexity and computing performance. The purpose of this study was to evaluate the accuracy of our newly-developed automatic 3D cephalometric annotation system using a deep learning algorithm. Our model system mainly consisted of a 3D convolutional neural network and image data resampling. Discrepancies between the referenced and predicted coordinate values in three axes and in 3D distance were calculated to yield prediction errors of 3.26, 3.18, and 4.81 mm (for three axes) and 7.61 mm (for 3D). Moreover, there was no difference (p > 0.05) among the landmarks of three groups (midsagittal plane, horizontal plane and mandible). Although our 3D convolutional neural network-based annotation system could not achieve the level of accuracy demanded by immediate clinical applications, it can nevertheless serve as an initial approximate guide to landmarks, thus reducing the time needed for annotation.
28 February 2022 in IEEE Access
Cone-beam computed tomography (CBCT) is widely used in dental and maxillofacial imaging applications. However, CBCT suffers from shading artifacts owing to several factors, including photon scattering and data truncation. This paper presents a deep-learning-based method for eliminating the shading artifacts that interfere with the diagnostic and treatment processes. The proposed method involves a two-stage generative adversarial network (GAN)-based image-to-image translation, where it operates on unpaired CBCT and multidetector computed tomography (MDCT) images. The first stage uses a generic GAN along with the fidelity difference between the original CBCT and MDCT-like images generated by the network. Although this approach is generally effective for denoising, at times, it introduces additional artifacts that appear as bone-like structures in the output images. This is because the weak input fidelity between the two imaging modalities can make it difficult to preserve the morphological structures from complex shadowing artifacts. The second stage of the proposed model addresses this problem. In this stage, paired training data, excluding inappropriate data, were collected from the results obtained in the first stage. Subsequently, the fidelity-embedded GAN was retrained using the selected paired samples. The results obtained in this study reveal that the proposed approach substantially reduces the shadowing and secondary artifacts arising from incorrect data fidelity while preserving the morphological structures of the original CBCT image. In addition, the corrected image obtained using the proposed method facilitates accurate bone segmentation compared to the original and corrected CBCT images obtained using the unpaired method.
28 August 2020 in IEEE Access
In low-dose computed tomography (LDCT), a penalized weighted least squares (PWLS) approach that incorporates the Poisson statistics of X-ray photons can significantly reduce excessive quantum noise. To improve the quality of LDCT images, prior information such as the total variation, Markov random field, and nonlocal mean, can be imposed onto the target image. However, this information may be limited to reflect the characteristics of the target images, thereby resulting in unexpected side effects (e.g. blurry images). In this paper, we propose a PWLS method combined with a deep learning prior, which is learned from standard dose CT (SDCT) images. The proposed model learns a noise reduction function that maps an LDCT image to its corresponding SDCT images and estimates the prior distribution using a Pearson χ² divergence. The model can be converted to the least squares generative adversarial network with an added PWLS objective, where the optimal generator acts as the noise reduction function. We reformulate the proposed model as a constrained optimization problem and solve it using the alternating optimization (AO) algorithm. Clinical SDCT and simulated LDCT scans of ten patients were used to show the validity of the proposed method. Results show that the proposed method outperforms other PWLS methods, by imposing priors such as the total variation and the nonlocal mean.
22 July 2024 in IEEE Access
This paper presents a fully automatic registration method of dental cone-beam computed tomography (CBCT) and face scan data. Difficulties in accurately merging facial scans and CBCT images arise from the different image acquisition methods and limited area of correspondence between the two facial surfaces. In addition, it is difficult to use machine learning techniques because they use face-related 3D medical data with radiation exposure, which are difficult to obtain for training. The proposed method addresses these problems by reusing an existing machine-learning-based 2D landmark detection algorithm in an open-source library and developing a novel mathematical algorithm that identifies paired 3D landmarks from knowledge of the corresponding 2D landmarks. A main contribution of this study is that the proposed method does not require annotated training data of facial landmarks because it uses a pre-trained facial landmark detection algorithm that is known to be robust and generalized to various 2D face image models. Note that this reduces a 3D landmark detection problem to a 2D problem of identifying the corresponding landmarks on two 2D projection images generated from two different projection angles. Here, the 3D landmarks for registration were selected from the sub-surfaces with the least geometric change under the CBCT and face scan environments. For the final fine-tuning of the registration, the Iterative Closest Point method was applied, which utilizes geometrical information around the 3D landmarks. The proposed method achieved an averaged mean surface distance error of 0.74 mm for three pairs of CBCT and face scan datasets.