Haruki Ueda

Dr. Haruki Ueda

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 May 2022 in BMC Musculoskeletal Disorders
Verified
Journal Article
Open access
PDF file
TITakuya Iimura
Takuya Iimura
HUHaruki Ueda
Haruki Ueda
SISatoshi Inami
Satoshi Inami
HMHiroshi Moridaira
Hiroshi Moridaira
DTDaisaku Takeuchi
Daisaku Takeuchi
HTHiroshi Taneichi
Hiroshi Taneichi
Background: The uniqueness of spinal sagittal alignment in thoracic adolescent idiopathic scoliosis (AIS), for example, the drastically smaller thoracic kyphosis seen in some patients, has been recognized but not yet fully understood. The purpose of this study was to clarify the characteristics of sagittal alignment of thoracic AIS and to determine the contributing factors. Methods: Whole spine radiographs of 83 thoracic AIS patients (73 females) were analyzed. The measured radiographic parameters were the Cobb angle of thoracic scoliosis, thoracic kyphosis (TK), lumbar lordosis (LL), C7 sagittal vertical axis (C7 SVA), pelvic incidence (PI), pelvic tilt (PT), and sacral slope (SS). Additionally, max-LL, which was defined as the maximum lordosis angle from the S1 endplate, the inflection point between thoracic kyphosis and lumbar lordosis, and the SVA of the inflection point (IP SVA) were measured. The factors significantly related to a decrease in TK were assessed by stepwise logistic regression analysis. In addition, cluster analysis was performed to classify the global sagittal alignment. Results: The significant factors for a decrease in TK were an increase in SS (p = 0.0003, [OR]: 1.16) and a decrease in max-LL (p = 0.0005, [OR]: 0.89). According to the cluster analysis, the global sagittal alignment was categorized into the following three types: Type 1 (low SS, low max-LL, n = 28); Type 2 (high SS, low max-LL, n = 22); and Type 3 (high SS, high max-LL, n = 33). Conclusions: In thoracic AIS, a decreased TK corresponded to an increased SS or a decreased max-LL. The sagittal alignment of thoracic AIS patients could be classified into three types based on SS and max-LL. One of these three types includes the unique sagittal profile of very small TK.

31 December 2021 in Healthcare
Verified
Journal Article
Open access
PDF file
HUHaruki Ueda
Haruki Ueda
HAHideyuki Arima
Hideyuki Arima
TKTokumi Kanemura
Tokumi Kanemura
MKMasao Koda
Masao Koda
MYMitsuru Yagi
Mitsuru Yagi
HTHiroshi Taneichi
Hiroshi Taneichi
(1) Background: Despite the number of complicated and expensive spine surgery procedures maintained by the national health insurance system in Japan, until now there has been no large-scale multicenter clinical database for this field to understand and improve healthcare expenditure and treatment outcomes. The purpose of this report is to announce the establishment and methodology of a nationwide registry system for spinal instrumentation surgeries by the Japanese Spinal Instrumentation Society (JSIS), and to report the progress over the first 1.5 years of this database’s operation. (2) Methods: The JSIS recently produced an online database with an electronic server. The collected information included patient background, surgery information, and early complications of primary and revision cases. Analysis included data from February 2018, when registration began, to August 2019. (3) Results: As of August 2019, 73 facilities have completed the required paperwork to start, and 55 facilities have registered cases. Of the total 5456 registered cases, 4852 were valid and 2511 were completed. (4) Conclusions: JSIS-DB, the nationwide web-based registry system for spinal instrumentation surgery in Japan, was launched for the purpose of research, healthcare policy regulation, and improved patient care, and its methodology and progress in the first 1.5 years are reported in this study.

13 October 2025 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
STShinji Takahashi
Shinji Takahashi
SIShota Ichikawa
Shota Ichikawa
KWKei Watanabe
Kei Watanabe
HUHaruki Ueda
Haruki Ueda
HAHideyuki Arima
Hideyuki Arima
Shoji SekiShoji Seki
Shoji Seki
Background/Objectives: The clinical management of adolescent idiopathic scoliosis (AIS) is hindered by the inability to accurately predict curve progression. Although skeletal maturity and the initial Cobb angle are established predictors of progression, their combined predictive accuracy remains limited. This study aimed to develop a robust and interpretable artificial intelligence (AI) system using deep learning (DL) models to predict the progression of scoliosis using only standing frontal radiographs. Methods: We conducted a multicenter study involving 542 patients with AIS. After excluding 52 borderline progression cases (6–9° progression in the Cobb angle), 294 and 196 patients were assigned to progression (≥10° increase) and non-progression (≤5° increase) groups, respectively, considering a 2-year follow-up. Frontal whole spinal radiographs were preprocessed using histogram equalization and divided into two regions of interest (ROIs) (ROI 1, skull base–femoral head; ROI 2, C7–iliac crest). Six pretrained DL models, including convolutional neural networks (CNNs) and transformer-based models, were trained on the radiograph images. Gradient-weighted class activation mapping (Grad-CAM) was further performed for model interpretation. Results: Ensemble models outperformed individual ones, with the average ensemble model achieving area under the curve (AUC) values of 0.769 for ROI 1 and 0.755 for ROI 2. Grad-CAM revealed that the CNNs tended to focus on the local curve apex, whereas the transformer-based models demonstrated global attention across the spine, ribs, and pelvis. Models trained on ROI 2 performed comparably with respect to those using ROI 1, supporting the feasibility of image standardization without a loss of accuracy. Conclusions: This study establishes the clinical potential of transformer-based DL models for predicting the progression of scoliosis using only plain radiographs. Our multicenter approach, high AUC values, and interpretable architectures support the integration of AI into clinical decision-making for the early treatment of AIS.

15 March 2025 in BMC Medical Education
Verified
Journal Article
Open access
PDF file
HUHaruki Ueda
Haruki Ueda
SISatoshi Inami
Satoshi Inami
HMHiroshi Moridaira
Hiroshi Moridaira
MTMasahiko Takahata
Masahiko Takahata
TITakuya Iimura
Takuya Iimura
HTHiroshi. Taneichi
Hiroshi. Taneichi
Advancements in understanding spinal disorders and diagnostic techniques have increased the range and complexity of spinal surgeries. However, constraints have arisen in gaining experience techniques through actual surgical cases due to considerations of medical safety, efficiency in working hours, and cost-effectiveness. As such, off-the-job training is expected to play an increasingly significant role. Three-dimensional models have been used for organizing knowledge and training surgical techniques. Their strengths lie in safety, accessibility, and cost-effectiveness. However, their proximity and limited realism make them less suitable for advanced training, restricting their use mainly to beginners. This study evaluates the potential of more realistic models for comprehensive surgical training and explores further applications of them. MRT2, a detailed three-dimensional model developed from real patient computed tomography (CT) data, offers realistic external characteristics and compatibility with radiological imaging. Three types of spine implant surgery seminars were conducted using MRT2: (A) cervical pedicle screw placement with fluoroscopy and navigation, (B) a comprehensive mock patient case—from understanding pathology to surgical planning and execution, and (C) lectures and hands-on deformity correction practice for various spinal conditions. Participants evaluated the models and seminars through questionnaires. Seminar A, comparing MRT2 to a conventional model (Sawbones), found MRT2 significantly more realistic in visual and performance realism, as well as tactile feedback (visual realism: MRT2 45, Sawbone 26, p = 0.0009; performance realism: MRT2 42, Sawbone 17, p = 0.0001; tactile feedback: MRT2 40, Sawbone 18, p = 0.009). In Seminar B, MRT2 provided an immersive experience even for spine surgery specialists, closely mimicking clinical practice (Questions 14–18 regarding psychological aspect, scoring 18–19 out of 20). Open-ended responses noted MRT2’s unique benefits, such as allowing multiple participants to perform the same procedure for comparative planning and outcomes. Observing vertebral movements during corrective maneuvers further confirmed its educational value. Enhanced structural detail and realistic simulation make these three-dimensional spinal models highly effective for both novice and specialist training, significantly improving the training experience across skill levels.

26 April 2024 in BMC Musculoskeletal Disorders
Verified
Journal Article
Open access
PDF file
HUHaruki Ueda
Haruki Ueda
TITakuya Iimura
Takuya Iimura
SISatoshi Inami
Satoshi Inami
HMHiroshi Moridaira
Hiroshi Moridaira
TYTakuya Yazawa
Takuya Yazawa
HTHiroshi Taneichi
Hiroshi Taneichi
Background: The natural history of the congenital spinal deformity and its clinical magnitude vary widely in human species. However, we previously reported that the spinal deformities of congenital scoliosis mice did not progress throughout our observational period according to soft X-ray and MRI data. In this study, congenital vertebral and intervertebral malformations in mice were assessed via magnetic resonance (MR) and histological images. Methods: Congenital spinal anomalies were chronologically assessed via soft X-ray and 7 T MR imaging. MR images were compared to the histological images to validate the findings around the malformations. Results: Soft X-ray images showed the gross alignment of the spine and the contour of the malformed vertebrae, with the growth plate and cortical bone visible as higher density lines, but could not be used to distinguish the existence of intervertebral structures. In contrast, MR images could be used to distinguish each structure, including the cortical bone, growth plate, cartilaginous end plate, and nucleus pulposus, by combining the signal changes on T1-weighted imaging (T1WI) and T2-weighted imaging (T2WI). The intervertebral structure adjacent to the malformed vertebrae also exhibited various abnormalities, such as growth plate and cartilaginous end plate irregularities, nucleus pulposus defects, and bone marrow formation. In the chronological observation, the thickness and shape of the malformed structures on T1WI did not change. Conclusions: Spinal malformations in mice were chronologically observed via 7 T MRI and histology. MR images could be used to distinguish the histological structures of normal and malformed mouse spines. Malformed vertebrae were accompanied by adjacent intervertebral structures that corresponded to the fully segmented structures observed in human congenital scoliosis, but the intervertebral conditions varied. This study suggested the importance of MRI and histological examinations of human congenital scoliosis patients with patterns other than nonsegmenting patterns, which may be used to predict the prognosis of patients with spinal deformities associated with malformed vertebrae.

27 December 2025 in Journal of Orthopaedic Science
Verified
Journal Article
Closed access
HUHaruki Ueda
Haruki Ueda
HAHideyuki Arima
Hideyuki Arima
KYKoji Yamada
Koji Yamada
MYMitsuru Yagi
Mitsuru Yagi
KYKimiaki Yokosuka
Kimiaki Yokosuka
AMAkira Matsumura
Akira Matsumura