RESUMEN

BACKGROUND: The accurate segmentation of spine muscles plays a crucial role in analyzing musculoskeletal disorders and designing effective rehabilitation strategies. Various imaging techniques such as MRI have been utilized to acquire muscle images, but the segmentation process remains complex and challenging due to the inherent complexity and variability of muscle structures. In this systematic review, we investigate and evaluate methods for automatic segmentation of spinal muscles. METHODS: Data for this study were obtained from PubMed/MEDLINE databases, employing a search methodology that includes the terms 'Segmentation spine muscle' within the title, abstract, and keywords to ensure a comprehensive and systematic compilation of relevant studies. Systematic reviews were not included in the study. RESULTS: Out of 369 related studies, we focused on 12 specific studies. All studies focused on segmentation of spine muscle use MRI, in this systematic review subjects such as healthy volunteers, back pain patients, ASD patient were included. MRI imaging was performed on devices from several manufacturers, including Siemens, GE. The study included automatic segmentation using AI, segmentation using PDFF, and segmentation using ROI. CONCLUSION: Despite advancements in spine muscle segmentation techniques, challenges still exist. The accuracy and precision of segmentation algorithms need to be improved to accurately delineate the different muscle structures in the spine. Robustness to variations in image quality, artifacts, and patient-specific characteristics is crucial for reliable segmentation results. Additionally, the availability of annotated datasets for training and validation purposes is essential for the development and evaluation of new segmentation algorithms. Future research should focus on addressing these challenges and developing more robust and accurate spine muscle segmentation techniques to enhance clinical assessment and treatment planning for musculoskeletal disorders.

Asunto(s)

RESUMEN

Background: Hip fractures are a common and debilitating condition, particularly among older adults. Loss of muscle mass and strength is a common consequence of hip fractures, which further contribute to functional decline and increased disability. Assessing changes in individual thigh muscles volume in follow-up patients can provide valuable insights into the quantitative recovery process and guide rehabilitation interventions. However, accurately measuring anatomical individual thigh muscle volume can be challenging due to various, labor intensive and time-consuming. Materials and Methods: This study aimed to evaluate differences in thigh muscle volume in followed-up hip fracture patients computed tomography (CT) scans using an AI based automatic muscle segmentation model. The study included a total of 18 patients at Gyeongsang National University, who had undergone surgical treatment for a hip fracture. We utilized the automatic segmentation algorithm which we have already developed using UNETR (U-net Transformer) architecture, performance dice score = 0.84, relative absolute volume difference 0.019 ± 0.017%. Results: The results revealed intertrochanteric fractures result in more significant muscle volume loss (females: -97.4 cm3, males: -178.2 cm3) compared to femoral neck fractures (females: -83 cm3, males: -147.2 cm3). Additionally, the study uncovered substantial disparities in the susceptibility to volume loss among specific thigh muscles, including the Vastus lateralis, Adductor longus and brevis, and Gluteus maximus, particularly in cases of intertrochanteric fractures. Conclusions: The use of an automatic muscle segmentation model based on deep learning algorithms enables efficient and accurate analysis of thigh muscle volume differences in followed up hip fracture patients. Our findings emphasize the significant muscle loss tied to sarcopenia, a critical condition among the elderly. Intertrochanteric fractures resulted in greater muscle volume deformities, especially in key muscle groups, across both genders. Notably, while most muscles exhibited volume reduction following hip fractures, the sartorius, vastus and gluteus groups demonstrated more significant disparities in individuals who sustained intertrochanteric fractures. This non-invasive approach provides valuable insights into the extent of muscle atrophy following hip fracture and can inform targeted rehabilitation interventions.

Asunto(s)

RESUMEN

Many older patients with COVID-19 likely have co-morbid osteoporosis. We investigated the clinical outcomes of COVID-19 patients with osteoporosis. This was a retrospective cohort study using national claims data from Korea encoded in the common data model. Patients aged ≥ 50 years diagnosed with COVID-19 infection between January 2020 and April 2022 were included and stratified into two groups according to a history of osteoporosis. Clinical outcomes of COVID-19 infection were analyzed using logistic regression analysis after large-scale propensity score stratification. Of the 597,011 patients with COVID-19 included in the study, 105,172 had a history of osteoporosis. In patients with a history of osteoporosis, the odds of mortality decreased (odds ratio [OR] 0.82, P < 0.002), whereas most clinical outcomes of COVID-19 did not exhibit differences compared to those without such a history. Osteoporosis patients with a history of fractures showed increased odds of pneumonia, hospitalization, major adverse cardiac events, venous thromboembolism, and mortality, compared to patients without osteoporosis (ORs 1.34-1.58, P < 0.001 to P = 0.001). Our study suggests that patients with severe osteoporosis who have experienced fractures have an elevated risk of severe complications with COVID-19, while osteoporosis patients without fractures who have sought medical attention have a lower risk of mortality.

Asunto(s)

RESUMEN

Background: Hemiarthroplasty is frequently used to treat displaced femoral neck fractures in elderly patients, but it has a higher risk of postoperative dislocation. We introduced the posterior approach and inferior capsulotomy (PAICO) to enhance joint stability after bipolar hemiarthroplasty for femoral neck fracture. We evaluated whether the PAICO would have a lower dislocation rate than the conventional posterior approach with superior capsulotomy. Methods: From January 2021 to December 2021, we prospectively recruited 25 patients (25 hips) aged 50 years or older who underwent bipolar hemiarthroplasty for femoral neck fractures due to low-energy trauma as the PAICO group. We compared the PAICO group with a historical control group who had undergone hemiarthroplasty in 7 institutes between 2010 and 2020. The primary endpoint was dislocation within 1 year after the surgery. We compared data from the PAICO group with the data from the historical control group from the Korean Hip Fracture Registry which was carried out in South Korea. Results: A total of 25 patients (25 hips) were enrolled in the present study; 3,477 patients (3,571 hips) who underwent bipolar hemiarthroplasty were reviewed as the historical control group. In the PAICO group, we observed no dislocation, whereas the dislocation rate in the control group was 1.3%. Conclusions: In patients with displaced femoral neck fractures, the PAICO approach demonstrated comparable results in operation time and complication rates when compared to bipolar hemiarthroplasty using superior capsulotomy. Notably, there were no observed cases of dislocation among patients who underwent the PAICO approach. We recommend this PAICO approach to surgeons using the posterior approach, hoping to prevent dislocation in bipolar hemiarthroplasty.

Asunto(s)

RESUMEN

BACKGROUND: As recognized by the World Health Organization in 2016 with its inclusion in the International Classification of Diseases, Tenth Revision as M62.84, and by South Korea in 2021 as M62.5, the diagnostic guidelines for sarcopenia vary globally. Despite its prevalence in older populations, data on sarcopenia in Koreans aged 60 and above is scarce, highlighting the need for research on its prevalence in this demographic. METHODS: Utilizing the 2022 Korea National Health and Nutrition Examination Survey dataset, sarcopenia was assessed among 1,946 individuals aged 60 or older according to the Asian Working Group for Sarcopenia 2019 criteria, incorporating grip strength and bioelectrical impedance analysis measurements. Statistical analyses were performed to differentiate categorical and continuous variables using logistic regression and Student's t-tests, respectively. RESULTS: The prevalence of sarcopenia was found to increase with age, with the highest prevalence observed in the oldest age group (80 years and older). The overall prevalence of sarcopenia in our study population was 6.8%. Among men, the prevalence of sarcopenia was 5.5% in the 60 or older age group, 9.6% in the 70 or older age group, and 21.5% in the 80 or older age group. Among women, the prevalence of sarcopenia was 7.9%, 10.5%, and 25.9%, respectively. CONCLUSIONS: This study highlights the significant burden of sarcopenia in elderly Koreans, particularly among the oldest individuals. These findings call for targeted interventions to manage and prevent sarcopenia, along with further research on its risk factors, consequences, and effective mitigation strategies.

RESUMEN

Introduction: Basicervical femoral neck fracture (FNF) is an uncommon type of femoral neck fracture and is associated with an increased risk of fixation failure due to its inherent instability. The purpose of this study was to compare the surgical parameters and reoperation rate between the use of a multiple cannulated screw (MCS) and fixed angle device (FAD) in treating basicervical FNFs. Methods: We retrospectively reviewed the records of 885 patients who underwent internal fixation between May 2004 and August 2019 to determine basicervical FNF with at least 12 months of follow-up. Among the identified 77 patients with basicervical FNF, 17 patients who underwent multiple cannulated screw (MCS) fixation and 36 patients who underwent fixed angle device (FAD) fixation were included. We compared the rates of fracture-site collapse and reoperations according to the fixation device. Results: Among the 53 patients with basicervical FNF, 13 patients (24.5%) sustained surgical complications (8 collapses of fracture site and 5 reoperations). The reoperation rate in the MCS group was significantly higher than that in the FAD group (23.5% vs. 2.8%, p = 0.016), without any significant difference in the collapse of the fracture site (11.8% vs. 16.7%, p = 0.642). Conclusions: Although basicervical FNF was rare among hip fractures, fracture site collapse was prevalent and prone to fixation failure. Surgeons should keep this in mind, and consider FAD for basicervical FNF.

Asunto(s)

RESUMEN

Templating is essential in hip arthroplasty preparation, facilitating implant size prediction and surgical rehearsal. It ensures the selection of suitable implants according to patient anatomy and disease, aiming to minimize post-operative complications. Various templating methods exist, including traditional acetate templating on both analog and digital images, alongside digital templating on digital images, which is categorized into 2D and 3D approaches. Despite the popularity of acetate templating on digital images, challenges such as the requirement for physical templates and result preservation persist. To address these limitations, digital templating with software like OrthoSize and Orthoview has been suggested, although not universally accessible. This technical note advocates for Microsoft PowerPoint as an effective alternative for 2D digital templating, highlighting its user-friendly features for image manipulation without needing specialized software. The described method involves scanning acetate templates, adjusting the images in PowerPoint 365 for size, position, and calibration on patient radiographs, and demonstrating reliability through preliminary assessments, with intraclass correlation coefficient (ICC) values indicating a high level of agreement for cup and stem size (ICC = 0.860, 0.841, respectively) but moderate for neck length (ICC = 0.592). We have introduced a method for performing 2D digital templating in the clinical field without the need for specialized software dedicated to digital templating. We believe this method significantly improves the accessibility to 2D digital templating, which was previously limited by the need for digital templating software. Additionally, it enables surgeons to easily establish arthroplasty plans and share them, overcoming the limitations of acetate templates.

RESUMEN

Background: As the population ages, the rates of hip diseases and fragility fractures are increasing, making total hip arthroplasty (THA) one of the best methods for treating elderly patients. With the increasing number of THA surgeries and diverse surgical methods, there is a need for standard evaluation protocols. This study aimed to use deep learning algorithms to classify THA videos and evaluate the accuracy of the labelling of these videos. Methods: In our study, we manually annotated 7 phases in THA, including skin incision, broaching, exposure of acetabulum, acetabular reaming, acetabular cup positioning, femoral stem insertion, and skin closure. Within each phase, a second trained annotator marked the beginning and end of instrument usages, such as the skin blade, forceps, Bovie, suction device, suture material, retractor, rasp, femoral stem, acetabular reamer, head trial, and real head. Results: In our study, we utilized YOLOv3 to collect 540 operating images of THA procedures and create a scene annotation model. The results of our study showed relatively high accuracy in the clear classification of surgical techniques such as skin incision and closure, broaching, acetabular reaming, and femoral stem insertion, with a mean average precision (mAP) of 0.75 or higher. Most of the equipment showed good accuracy of mAP 0.7 or higher, except for the suction device, suture material, and retractor. Conclusions: Scene annotation for the instrument and phases in THA using deep learning techniques may provide potentially useful tools for subsequent documentation, assessment of skills, and feedback.

Asunto(s)

RESUMEN

The study aims to develop a deep learning based automatic segmentation approach using the UNETR(U-net Transformer) architecture to quantify the volume of individual thigh muscles(27 muscles in 5 groups) for Sarcopenia assessment. By automating the segmentation process, this approach improves the efficiency and accuracy of muscle volume calculation, facilitating a comprehensive understanding of muscle composition and its relationship to Sarcopenia. The study utilized a dataset of 72 whole thigh CT scans from hip fracture patients, annotated by two radiologists. The UNETR model was trained to perform precise voxel-level segmentation and various metrics such as dice score, average symmetric surface distance, volume correlation, relative absolute volume difference and Hausdorff distance were employed to evaluate the model's performance. Additionally, the correlation between Sarcopenia and individual thigh muscle volumes was examined. The proposed model demonstrated superior segmentation performance compared to the baseline model, achieving higher dice scores (DC = 0.84) and lower average symmetric surface distances (ASSD = 1.4191 ± 0.91). The volume correlation between Sarcopenia and individual thigh muscles in the male group. Furthermore, the correlation analysis of grouped thigh muscles also showed negative associations with Sarcopenia in the male participants. This thesis presents a deep learning based automatic segmentation approach for quantifying individual thigh muscle volume in sarcopenia assessment. The results highlights the associations between Sarcopenia and specific individual muscles as well as grouped thigh muscle regions, particularly in males. The proposed method improves the efficiency and accuracy of muscle volume calculation, contributing to a comprehensive evaluation of Sarcopenia. This research enhances our understanding of muscle composition and performance, providing valuable insights for effective interventions in Sarcopenia management.

Asunto(s)

RESUMEN

OBJECTIVE: Patients with Parkinson's disease (PD) have an increased risk of sarcopenia which is expected to negatively affect gait, leading to poor clinical outcomes including falls. In this study, we investigated the gait patterns of patients with PD with and without sarcopenia (sarcopenia and non-sarcopenia groups, respectively) using an app-derived program and explored if gait parameters could be utilized to predict sarcopenia based on machine learning. METHODS: Clinical and sarcopenia profiles were collected from patients with PD at Hoehn and Yahr (HY) stage ≤ 2. Sarcopenia was defined based on the updated criteria of the Asian Working Group for Sarcopenia. The gait patterns of the patients with and without sarcopenia were recorded and analyzed using a smartphone application. The random forest model was applied to predict sarcopenia in patients with PD. RESULTS: Data from 38 patients with PD were obtained, among which 9 (23.7%) were with sarcopenia. Clinical parameters were comparable between the sarcopenia and non-sarcopenia groups. Among various clinical and gait parameters, the average range of motion of the hip joint showed the highest association with sarcopenia. Based on the random forest algorithm, the combined difference in knee and ankle angles from standing still before walking to the maximum angle during walking (Kneeankle_diff), the difference between the angle when standing still before walking and the maximum angle during walking for the ankle (Ankle_dif), and the min angle of the hip joint (Hip_min) were the top three features that best predict sarcopenia. The accuracy of this model was 0.949. CONCLUSIONS: Using smartphone app and machine learning technique, our study revealed gait parameters that are associated with sarcopenia and that help predict sarcopenia in PD. Our study showed potential application of advanced technology in clinical research.

Asunto(s)

RESUMEN

Age-related sarcopenia, resulting from a gradual loss in skeletal muscle mass and strength, is pivotal to the increased prevalence of functional limitation among the older adult community. The purpose of this meta-analysis of individual patient data is to investigate the difference in health-related quality of life between sarcopenic individuals and those without the condition using the Sarcopenia Quality of Life (SarQoL) questionnaire. A protocol was published on PROSPERO. Multiple databases and the grey literature were searched until March 2023 for studies reporting quality of life assessed with the SarQoL for patients with and without sarcopenia. Two researchers conducted the systematic review independently. A two-stage meta-analysis was performed. First, crude (mean difference) and adjusted (beta coefficient) effect sizes were calculated within each database; then, a random effect meta-analysis was applied to pool them. Heterogeneity was measured using the Q-test and I2 value. Subgroup analyses were performed to investigate the source of potential heterogeneity. The strength of evidence of this association was assessed using GRADE. From the 413 studies identified, 32 were eventually included, of which 10 were unpublished data studies. Sarcopenic participants displayed significantly reduced health-related quality of life compared with non-sarcopenic individuals (mean difference = -12.32; 95 % CI = [-15.27; -9.37]). The model revealed significant heterogeneity. Subgroup analyses revealed a substantial impact of regions, clinical settings, and diagnostic criteria on the difference in health-related quality of life between sarcopenic and non-sarcopenic individuals. The level of evidence was moderate. This meta-analysis of individual patient data suggested that sarcopenia is associated with lower health-related quality of life measured with SarQoL.

Asunto(s)

RESUMEN

BACKGROUND: This study aimed to investigate real-world data of C-terminal telopeptide (CTX), propeptide of type I collagen (P1NP), and osteocalcin through present multicenter clinical study, and retrospectively analyze the usefulness of bone turnover markers (BTMs) in Koreans. METHODS: The study focused on pre- and post-menopausal patients diagnosed with osteoporosis and excluded patients without certain test results or with test intervals of over 1 year. The demographic data and 3 BTMs (CTX, P1NP, and osteocalcin) were collected. The patients were classified by demographic characteristics and the BTM concentrations were analyzed by the group. RESULTS: Among women with no history of fractures, the levels of P1NP (N=2,100) were 43.544±36.902, CTX (N=1,855) were 0.373 ±0.927, and osteocalcin (N=219) were 10.81 ±20.631. Among men with no history of fractures, the levels of P1NP (N=221) were 48.498±52.892, CTX (N=201) were 0.370±0.351, and osteocalcin (N=15) were 7.868 ±10.674. Treatment with teriparatide increased the P1NP levels after 3 months in both men and women, with a 50% increase observed in women. Similarly, treatment with denosumab decreased the CTX levels after 3 months in both men and women, with a reduction of 50% observed in women. CONCLUSIONS: The results of this study can contribute to the accurate assessment of bone replacement status in Koreans. We also provide the P1NP level in the Korean population for future comparative studies with other populations.

RESUMEN

Background: Increasing longevity has caused the very old population to become the fastest-growing segment. The number of centenarians (over 100 years old) is increasing rapidly. Fractures in the elderly lead to excessive medical costs and decreased quality of life with socioeconomic burdens. However, little research has thoroughly examined the functional outcomes and mortality of hip fractures in centenarians. Methods: This is a retrospective observational study. Sixty-eight centenarian hip fracture patients were admitted to the 10 institutions from February 2004 to December 2019. Fifty-six patients with 1-year follow-up were finally included. The following data were obtained: sex, age, body mass index, Charlson comorbidity index value on the operation day, Koval's classification for ambulatory ability, type of fracture, the time interval from trauma to surgery, American Society of Anesthesiologists grade, surgery-related complications, and duration of hospital stay. Postoperative Koval's classification (at 1 year after surgery) and information about death were also collected. Multivariate analysis was performed to analyze the risk factors affecting mortality 1 year after surgery. Results: Mortality rates were 26.8% at 6 months and 39.3% at 1 year. The 90-day mortality was 19.6%, and one of them (2.1%) died in the hospital. The 1-year mortality rates for the community ambulatory and non-community ambulatory groups were 29% and 52%, respectively. Only 9 (16.1%) were able to walk outdoors 1 year after surgery. The remaining 47 patients (83.9%) had to stay indoors after surgery. Multivariate analysis demonstrated that the pre-injury ambulatory level (adjusted hazard ratio, 2.884; p = 0.034) was associated with the risk of mortality. Conclusions: We report a 1-year mortality rate of 39.3% in centenarian patients with hip fractures. The risk factor for mortality was the pre-injury ambulatory status. This could be an important consideration in the planning of treatment for centenarian hip fracture patients.

Asunto(s)

RESUMEN

Background and Objectives: Hip fractures are commonly found in elderly patients, and often result in chronic pain and decreased physical function, as well as worsening of overall health. It is known that early surgical intervention during the acute phase and rehabilitation are important for improving clinical outcomes for these patients. However, the importance of management for improving the quality of life of these patients is becoming more emphasized. Studies on changes in sleep patterns after hip fractures are rare overseas. Therefore, the aim of this study is to investigate the prevalence of sleep disturbance in patients with hip fractures and to analyze the changes in sleep disturbance after surgery by comparing the preoperative and postoperative results. Materials and Methods: During the period from August 2022 to January 2023, patients who underwent surgical treatment for hip fractures and were recruited into the REAL Hip Cohort were selected as research subjects. The sleep survey was conducted using the Pittsburgh Sleep Quality Index (PSQI). The PSQI is composed of 18 questions, each divided into areas of sleep quality, sleep latency, duration, efficiency, disturbance, use of medication, and daytime dysfunction. Each area is scored 0-3 points and the total is 0-21. A score greater than five indicates sleep disorder. The PSQI was surveyed during hospitalization and three months after surgery for post-fracture sleep status. To analyze changes before and after the fracture, paired T-tests and chi-square tests were performed. Results: From August 2022 to January 2023, a total of 40 patients who were recruited into the REAL Hip Cohort responded to the PSQI survey. The average age was 77.4 years and 36 were female. Sleep quality worsened from 0.75 ± 1.0 before surgery to 1.4 ± 1.0 three months after surgery (p = 0.019), and sleep efficiency also worsened from 0.4 ± 0.6 to 1.4 ± 1.0 (p < 0.001). The PSQI increased from an average of 5.2 ± 2.8 before surgery to 8.2 ± 4.2 three months after surgery (p = 0.007), and the number of patients who could be diagnosed with sleep disorders also increased from 12 (40%) to 24 (60%) (p = 0.030). Conclusions: A decline in overall sleep status was observed in patients in a survey on sleep patterns three months after hip fracture. Additional management is needed to improve their sleep patterns.

Asunto(s)

RESUMEN

Periprosthetic joint infection (PJI) can be diagnosed to characterize the microorganisms constituting a biofilm, which is an essential procedure for proper treatment. The gold standard method for detecting and identifying the causative microorganism is culture of microorganisms from patients-derived sample.; however, this method takes a long time and has low sensitivity. To compensate for these limitations, identification methods based on real-time PCR (RT-PCR) have been widely used. However, RT-PCR also has limitations, including low sensitivity and the requirement of a standard curve for quantification. Therefore, to prevent significant proliferation of pathogenic bacteria, it is important to detect a limited number of infectious bacteria during early stages of PJI. In the present study, we developed droplet digital PCR-based detection of bacterial pathogens in PJI. And we evaluated the analytical performance of the assay using a model plasmid, based on the 16S ribosomal DNA sequence of target bacteria commonly found in PJI. We also prepared genomic DNA extracted from E. coli, S. aureus, and S. epidermidis to test whether ddPCR provides better sensitivity and quantification of the target sequences. ddPCR detected 400 attograms of target DNA, which was more than 10 times less than that detected by real-time PCR using synthesized plasmid. In addition, ddPCR detected target regions from genomic DNA of 50 femtograms for E. coli, 70 femtograms for S. epidermidis, and 90 femtograms for S. aureus. The results indicate that ddPCR has the potential to decrease the microbial detection limit and provide precise detection, signifying its effectiveness for early PJI.

Asunto(s)

RESUMEN

BACKGROUND: The relationship between physical function, musculoskeletal disorders and sarcopenia is intricate. Current physical function tests, such as the gait speed test and the chair stand test, have limitations in eliminating subjective influences. To overcome this, smart devices utilizing inertial measurement unit sensors and artificial intelligence (AI)-based methods are being developed. METHODS: We employed cutting-edge technologies, including the smart insole device and pose estimation based on AI, along with three classification models: random forest (RF), support vector machine and artificial neural network, to classify control and sarcopenia groups. Patient data of 83 individuals were divided into train and test sets, with approximately 67% allocated for training. Classification models were implemented using RStudio, considering individual and combined variables obtained through pose estimation and smart insole measurements. RESULTS: Performance evaluation of the classification models utilized accuracy, precision, recall and F1-score indicators. Using only pose estimation variables, accuracy ranged from 0.92 to 0.96, with F1-scores of 0.94-0.97. Key variables identified by the RF model were 'Hip_dif', 'Ankle_dif' and 'Hipankle_dif'. Combining variables from both methods increased accuracy to 0.80-1.00, with F1-scores of 0.73-1.00. CONCLUSIONS: In our study, a classification model that integrates smart insole and pose estimation technology was assessed. The RF model showed impressive results, particularly in the case of the Hip and Ankle variables. The growth of advanced measurement technologies suggests a promising avenue for identifying and utilizing additional digital biomarkers in the management of various disorders. The convergence of AI technologies with diagnostics and treatment approaches a promising future for enhanced interventions in conditions like sarcopenia.

Asunto(s)

RESUMEN

Osteoporosis is a health condition that involves weak bone mass and a deteriorated microstructure, which consequently lead to an increased risk of bone fractures with age. In elderly people, a fracture attributable to osteoporosis elevates mortality. The objective of this review was to examine the effects of macronutrients on bone mineral density (BMD), bone turnover markers (BTMs), and bone fracture in elderly people based on human studies. A systematic search was conducted in the PubMed®/MEDLINE® database. We included human studies published up to April 2023 that investigated the association between macronutrient intake and bone health outcomes. A total of 11 meta-analyses and 127 individual human studies were included after screening the records. Carbohydrate consumption seemed to have neutral effects on bone fracture in limited studies, but human studies on carbohydrates' effects on BMD or/and BTMs are needed. The human studies analyzed herein did not clearly show whether the intake of animal, vegetable, soy, or milk basic proteins has beneficial effects on bone health due to inconsistent results. Moreover, several individual human studies indicated an association between eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA), and osteocalcin. Further studies are required to draw a clear association between macronutrients and bone health in elderly people.

Asunto(s)

RESUMEN

BACKGROUND: Dual energy X-ray absorptiometry (DXA) is a preferred modality for screening or diagnosis of osteoporosis and can predict the risk of hip fracture. However, the DXA test is difficult to implement easily in some developing countries, and fractures have been observed before patients underwent DXA. The purpose of this systematic review is to search for studies that predict the risk of hip fracture using artificial intelligence (AI) or machine learning, organize the results of each study, and analyze the usefulness of this technology. METHODS: The PubMed, OVID Medline, Cochrane Collaboration Library, Web of Science, EMBASE, and AHRQ databases were searched including "hip fractures" AND "artificial intelligence". RESULTS: A total of 7 studies are included in this study. The total number of subjects included in the 7 studies was 330,099. There were 3 studies that included only women, and 4 studies included both men and women. One study conducted AI training after 1:1 matching between fractured and non-fractured patients. The area under the curve of AI prediction model for hip fracture risk was 0.39 to 0.96. The accuracy of AI prediction model for hip fracture risk was 70.26% to 90%. CONCLUSIONS: We believe that predicting the risk of hip fracture by the AI model will help select patients with high fracture risk among osteoporosis patients. However, to apply the AI model to the prediction of hip fracture risk in clinical situations, it is necessary to identify the characteristics of the dataset and AI model and use it after performing appropriate validation.

RESUMEN

BACKGROUND: The purpose of this study was to verify the accuracy and validity of using machine learning (ML) to select risk factors, to discriminate differences in feature selection by ML between men and women, and to develop predictive models for patients with osteoporosis in a big database. METHODS: The data on 968 observed features from a total of 3,484 the Korea National Health and Nutrition Examination Survey participants were collected. To find preliminary features that were well-related to osteoporosis, logistic regression, random forest, gradient boosting, adaptive boosting, and support vector machine were used. RESULTS: In osteoporosis feature selection by 5 ML models in this study, the most selected variables as risk factors in men and women were body mass index, monthly alcohol consumption, and dietary surveys. However, differences between men and women in osteoporosis feature selection by ML models were age, smoking, and blood glucose level. The receiver operating characteristic (ROC) analysis revealed that the area under the ROC curve for each ML model was not significantly different for either gender. CONCLUSIONS: ML performed a feature selection of osteoporosis, considering hidden differences between men and women. The present study considers the preprocessing of input data and the feature selection process as well as the ML technique to be important factors for the accuracy of the osteoporosis prediction model.