Your browser doesn't support javascript.
loading
Show: 20 | 50 | 100
Results 1 - 20 de 373
Filter
1.
Article in English | WPRIM | ID: wpr-787135

ABSTRACT

Although microscopic analysis of tissue slides has been the basis for disease diagnosis for decades, intra- and inter-observer variabilities remain issues to be resolved. The recent introduction of digital scanners has allowed for using deep learning in the analysis of tissue images because many whole slide images (WSIs) are accessible to researchers. In the present study, we investigated the possibility of a deep learning-based, fully automated, computer-aided diagnosis system with WSIs from a stomach adenocarcinoma dataset. Three different convolutional neural network architectures were tested to determine the better architecture for tissue classifier. Each network was trained to classify small tissue patches into normal or tumor. Based on the patch-level classification, tumor probability heatmaps can be overlaid on tissue images. We observed three different tissue patterns, including clear normal, clear tumor and ambiguous cases. We suggest that longer inspection time can be assigned to ambiguous cases compared to clear normal cases, increasing the accuracy and efficiency of histopathologic diagnosis by pre-evaluating the status of the WSIs. When the classifier was tested with completely different WSI dataset, the performance was not optimal because of the different tissue preparation quality. By including a small amount of data from the new dataset for training, the performance for the new dataset was much enhanced. These results indicated that WSI dataset should include tissues prepared from many different preparation conditions to construct a generalized tissue classifier. Thus, multi-national/multi-center dataset should be built for the application of deep learning in the real world medical practice.


Subject(s)
Adenocarcinoma , Classification , Dataset , Diagnosis , Learning , Observer Variation , Stomach
2.
Article in English | WPRIM | ID: wpr-786084

ABSTRACT

Post-transcriptional regulations of mRNA transcripts such as alternative splicing and alternative polyadenylation can affect the expression of genes without changing the transcript levels. Recent studies have demonstrated that these post-transcriptional events can have significant physiological impacts on various biological systems and play important roles in the pathogenesis of a number of diseases, including cancers. Nevertheless, how cellular signaling pathways control these post-transcriptional processes in cells are not very well explored in the field yet. The mammalian target of rapamycin complex 1 (mTORC1) pathway plays a key role in sensing cellular nutrient and energy status and regulating the proliferation and growth of cells by controlling various anabolic and catabolic processes. Dysregulation of mTORC1 pathway can tip the metabolic balance of cells and is associated with a number of pathological conditions, including various types of cancers, diabetes, and cardiovascular diseases. Numerous reports have shown that mTORC1 controls its downstream pathways through translational and/or transcriptional regulation of the expression of key downstream effectors. And, recent studies have also shown that mTORC1 can control downstream pathways via post-transcriptional regulations. In this review, we will discuss the roles of post-transcriptional processes in gene expression regulations and how mTORC1-mediated post-transcriptional regulations contribute to cellular physiological changes. We highlight post-transcriptional regulation as an additional layer of gene expression control by mTORC1 to steer cellular biology. These emphasize the importance of studying post-transcriptional events in transcriptome datasets for gaining a fuller understanding of gene expression regulations in the biological systems of interest.


Subject(s)
Alternative Splicing , Cardiovascular Diseases , Dataset , Gene Expression , Polyadenylation , RNA, Messenger , Sirolimus , Social Control, Formal , Transcriptome
3.
Article in English | WPRIM | ID: wpr-782505

ABSTRACT

10% of labeled tumor cells) of TNF receptor 1 (TNFR1), the protein product of TNFRSF1A gene, was correlated with sarcomatoid dedifferentiation and was an independent predictive factor of clinically unfavorable response and shorter survivals in separated TKI-treated ccRCC cohort.CONCLUSION: TNF-α signaling may play a role in TKI resistance, and TNFR1 expression may serve as a predictive biomarker for clinically unfavorable TKI responses in ccRCC.


Subject(s)
Biomarkers , Carcinoma, Renal Cell , Cohort Studies , Dataset , Drug Resistance , Gene Expression , Gene Expression Profiling , Heterografts , Humans , Immunohistochemistry , Protein-Tyrosine Kinases , Receptors, Tumor Necrosis Factor , Receptors, Tumor Necrosis Factor, Type I , Tumor Necrosis Factor-alpha
10.
Article in Korean | WPRIM | ID: wpr-811316

ABSTRACT

PURPOSE: We determined whether elevated serum alkaline phosphatase (ALP) was related to prevalence, location, type, length, and recurrence of pterygium in a population from the Republic of Korea.METHODS: A nationwide cross-sectional dataset, the Korean National Health and Nutrition Examination Survey (2008–2011), was used in this study. All participants were > 30 years of age and underwent the ALP test and ophthalmic evaluation (n = 22,359). One-way analysis of variance, the chi-square test, and Fisher's exact test were used to compare characteristics and outcomes among participants. Multivariable logistic regression was used to examine the possible associations between serum ALP levels and various types of pterygium. Data were adjusted for known risk factors for development of pterygium and ALP elevation (age, sex, residence, sunlight exposure, drinking, smoking, hypertension, diabetes, BMI, AST, ALT, vitamin D, and HDL).RESULTS: The overall prevalence of pterygium was 8.1%, and participants with pterygium had higher levels of serum ALP (p < 0.001). Participants with higher serum ALP had a significantly higher prevalence of all types of pterygium than those in the lower serum ALP quartiles. After adjusting for potential confounding factors, multivariate logistic regression analysis revealed that ALP was associated with the prevalence of pterygium (odds ratio [OR], 1.001; p = 0.038). Trend analysis between the OR and ALP quartiles revealed a linear trend in overall prevalence and in the intermediate type of pterygium. Subgroup analysis revealed a stronger correlation in participants > 50 years of age. One-way analysis of variance revealed an association between the size of pterygium and serum ALP quartile levels. Serum ALP was not associated with recurrence of pterygium.CONCLUSIONS: Increased serum ALP was associated with the prevalence and size of pterygium.


Subject(s)
Alkaline Phosphatase , Cross-Sectional Studies , Dataset , Drinking , Hypertension , Korea , Logistic Models , Nutrition Surveys , Prevalence , Pterygium , Recurrence , Republic of Korea , Risk Factors , Smoke , Smoking , Sunlight , Vitamin D
11.
Article in English | WPRIM | ID: wpr-811219

ABSTRACT

OBJECTIVE: To investigate pathologic discrepancies between colposcopy-directed biopsy (CDB) of the cervix and loop electrosurgical excision procedure (LEEP) in women with cytologic high-grade squamous intraepithelial lesions (HSILs).METHODS: We retrospectively identified 297 patients who underwent both CDB and LEEP for HSILs in cervical cytology between 2015 and 2018, and compared their pathologic results. Considering the LEEP to be the gold standard, we evaluated the diagnostic performance of CDB for identifying cervical intraepithelial neoplasia (CIN) grades 2 and 3, adenocarcinoma in situ, and cancer (HSIL+). We also performed age subgroup analyses.RESULTS: Among the study population, 90.9% (270/297) had pathologic HSIL+ using the LEEP. The diagnostic performance of CDB for identifying HSIL+ was as follows: sensitivity, 87.8%; specificity, 59.3%; balanced accuracy, 73.6%; positive predictive value, 95.6%; and negative predictive value, 32.7%. Thirty-three false negative cases of CDB included CIN2,3 (n=29) and cervical cancer (n=4). The pathologic HSIL+ rate in patients with HSIL− by CDB was 67.3% (33/49). CDB exhibited a significant difference in the diagnosis of HSIL+ compared to LEEP in all patients (p<0.001). In age subgroup analyses, age groups <35 years and 35–50 years showed good agreement with the entire data set (p=0.496 and p=0.406, respectively), while age group ≥50 years did not (p=0.036).CONCLUSION: A significant pathologic discrepancy was observed between CDB and LEEP results in women with cytologic HSILs. The diagnostic inaccuracy of CDB increased in those ≥50 years of age.


Subject(s)
Adenocarcinoma in Situ , Biopsy , Cervical Intraepithelial Neoplasia , Cervix Uteri , Colposcopy , Conization , Dataset , Diagnosis , Early Detection of Cancer , Female , Humans , Papanicolaou Test , Retrospective Studies , Sensitivity and Specificity , Squamous Intraepithelial Lesions of the Cervix , Uterine Cervical Neoplasms
12.
Annals of Dermatology ; : 130-140, 2020.
Article in English | WPRIM | ID: wpr-811085

ABSTRACT

BACKGROUND: Atopic dermatitis (AD) is recognized as a common inflammatory skin disease and frequently occurred in Asian and Black individuals.OBJECTIVE: Since the limitation of dataset associated with human severe AD, this study aimed to screen potential novel biomarkers involved in mild AD.METHODS: Expression profile data (GSE75890) were obtained from the database of Gene Expression Omnibus. Using limma package, the differentially expressed genes (DEGs) between samples from AD and healthy control were selected. Furthermore, function analysis was conducted. Meanwhile, the protein-protein interaction (PPI) network and transcription factor (TF)-miRNA-target regulatory network were constructed. And quantitative real-time polymerase chain reaction (qRT-PCR) was used to validate the expressions patterns of key genes.RESULTS: In total, 285 DEGs including 214 upregulated and 71 downregulated genes were identified between samples from two groups. The upregulated DEGs were mainly involved in nine pathways, such as hematopoietic cell lineage, pertussis, p53 signaling pathway, staphylococcus aureus infection, and cell cycle, while tight junction was the only pathway enriched by the downregulated DEGs. Cyclin B (CCNB)1, CCNB2, cyclin A (CCNA)2, C-X-C motif chemokine ligand (CXCL)10, and CXCL9 were key nodes in PPI network. The TF-miRNA-target gene regulatory network focused on miRNAs such as miR-106b, miR-106a, and miR-17, TFs such as nuclear factor kappa B subunit 1, RELA proto-oncogene, Sp1 transcription factor, and genes such as matrix metallopeptidase 9, peroxisome proliferator activated receptor gamma , and serpin family E member 1. Moreover, the upregulation of these genes, including CCNB1, CCNB2, CCNA2, CXCL10, and CXCL9 were confirmed by qRT-PCR.CONCLUSION: CCNB1, CCNB2, CCNA2, and CXCL9 might be novel markers of mild AD. miR-106b and miR-17 may involve in regulation of immune response in AD patients.


Subject(s)
Asian Continental Ancestry Group , Biomarkers , Cell Cycle , Cell Lineage , Computational Biology , Cyclin A , Cyclin B , Dataset , Dermatitis , Dermatitis, Atopic , Gene Expression , Gene Regulatory Networks , Humans , MicroRNAs , NF-kappa B , PPAR gamma , Proto-Oncogenes , Real-Time Polymerase Chain Reaction , Skin Diseases , Sp1 Transcription Factor , Staphylococcus aureus , Tight Junctions , Transcription Factors , Up-Regulation , Whooping Cough
13.
Article in English | WPRIM | ID: wpr-719310

ABSTRACT

OBJECTIVE: Elderly age is one of the poor prognostic factors in epithelial ovarian cancer (EOC), but the optimal age cut-off is not known. The present study sought to identify the ideal age cutoff that represents a negative prognostic factor in EOC, considering the geriatric assessment. METHODS: Hazard ratios (HRs) with p-values were calculated using all possible age cutoffs with stage, histology, grade, optimality and comorbidities as covariates in multivariate Cox regression model. The trends of p-value and HR by age cutoff were further evaluated in a subgroup of histology and in The Cancer Genome Atlas (TCGA) dataset. In addition, propensity score-matching analysis using the identified age cutoff was performed. RESULTS: An age of 66 years was shown to be the most significant cutoff for defining old age with independent prognostic power (HR=1.45; 95% confidence interval=1.04–2.03; p=0.027). This result was also observed with the analyses of serous histology subgroup and with the analysis of a TCGA dataset with serous EOC. In survival analysis, patients aged ≥66 years had significantly worse overall survival compared with younger individuals (56 months vs. 87 months; p=0.006), even following propensity score matching (57 vs. 78 months; p=0.038). CONCLUSION: An age of 66 years is the best cutoff to define elderly age in serous EOC patients considering the geriatric assessment, and this information can be used in the administration of individualized therapies in elderly EOC patients.


Subject(s)
Aged , Comorbidity , Dataset , Genome , Geriatric Assessment , Humans , Ovarian Neoplasms , Prognosis , Propensity Score
14.
Article in English | WPRIM | ID: wpr-719309

ABSTRACT

OBJECTIVE: To develop and validate a 3-year recurrence prediction score (RPS) system for predicting the baseline risk of recurrence of stage I–II endometrial carcinoma. METHODS: We reviewed 427 patients with International Federation of Gynecology and Obstetrics staging I–II endometrial carcinoma underwent surgery without any adjuvant therapy from 2005 to 2013. The patients were divided into 2 groups: the test cohort (n=251) comprising those who underwent surgery in odd-numbered years, and the validation cohort (n=176) comprising those who underwent surgery in even-numbered years. Multivariate analysis was performed using 7 candidate predictors to identify the risk factors for 3-year recurrence-free interval (RFI) in the test cohort. Each risk factor was scored based on logistic regression analyses of the test data set, and the sum of the risk factor scores was defined as the RPS system. We then applied the system in the validation cohort. RESULTS: Multivariate analysis revealed that the significant risk factors were age ≥60 years, pathological type II, positive cervical stromal invasion, and positive peritoneal cytology. In the test cohort, the 3-year RFI rates were 100%, 95.8%, 79.9%, and 33.3% for RPSs of 0, 1, 2, and 3, respectively. In the validation cohort, the 3-year RFI was significantly higher in the low-RPS group (RPS 0 or 1) than in the high-RPS group (RPS 2 or 3) (95.2% vs. 79.9%, p < 0.01). CONCLUSIONS: The RPS system shows significant reproducibility for predicting the baseline risk of recurrence. The system could potentially impact the choice of adjuvant therapy for stage I–II endometrial carcinoma.


Subject(s)
Cohort Studies , Dataset , Endometrial Neoplasms , Female , Gynecology , Humans , Logistic Models , Multivariate Analysis , Obstetrics , Recurrence , Risk Factors
15.
Article in English | WPRIM | ID: wpr-719606

ABSTRACT

PURPOSE: The aim of this study was to identify the minimally meaningful dosage of inulin leading to a prebiotic effect in Indonesian infants. METHODS: In a randomized controlled double-blinded, parallel, 3-arm intervention study, 164 healthy formula-fed infants aged 3 to 5 months first obtained formula-A (without inulin) during a 4-week adaptation period. Subsequently, 142 subjects were subjected to a 4-week feeding period by administering either formula-A (no inulin), formula-B (0.2 g/100 mL inulin) or formula-C (0.4 g/100 mL inulin). The primary outcome parameter was %-bifidobacteria in faecal samples determined using quantitative polymerase chain reaction analyses. Secondary outcome parameters were faecal %-lactobacilli, pH and stool frequency, and consistency. Growth and tolerance/adverse effects were recorded as safety parameters. RESULTS: Typical %-bifidobacteria and %-lactobacilli at the end of the adaptation period in the study population were 14% and 2%, respectively. For faecal pH, significant differences between formula groups A vs. C and A vs. B were found at the end of the intervention period. Testing for differences in faecal %-bifidobacteria and %-lactobacilli between groups was hampered by non-normal data set distributions; no statistically significant differences were obtained. Comparisons within groups revealed that only in formula group C, all the three relevant parameters exhibited a significant effect with an increase in faecal %-bifidobacteria and %-lactobacilli and a decrease in pH. CONCLUSION: A consistent prebiotic effect along with a decrease in pH and increase in %-bifidobacteria and %-lactobacilli was found only in the group administered 0.4 g inulin/100 mL.


Subject(s)
Dataset , Gastrointestinal Microbiome , Humans , Hydrogen-Ion Concentration , Infant Formula , Infant , Inulin , Polymerase Chain Reaction , Prebiotics
16.
Article in English | WPRIM | ID: wpr-719508

ABSTRACT

BACKGROUND: Elderly asthma (EA) is increasing, but the pathogenesis is unclear. This study aimed to identify EA-related biological pathways by analyzing genome-wide gene expression profiles in sputum cells. METHODS: A total of 3,156 gene probes with significantly differential expressions between EA and healthy elderly controls were used for a hierarchical clustering of genes to identify gene clusters. Gene set enrichment analysis provided biological information, with replication from Gene Expression Omnibus expression profiles. RESULTS: Fifty-five EA patients and 10 elderly control subjects were enrolled. Two distinct gene clusters were found. Cluster 1 (n = 35) showed a lower eosinophil proportion in sputum and less severe airway obstruction compared to cluster 2 (n = 20). The replication data set also identified 2 gene clusters (clusters 1' and 2'). Among 5 gene sets significantly enriched in cluster 1 and 3 gene sets significantly enriched in cluster 2, we confirmed that 2 were significantly enriched in the replication data set (OXIDATIVE_PHOSPHORYLATION gene set in cluster 1 and EPITHELIAL MESENCHYMAL TRANSITION gene set in cluster 2'). CONCLUSIONS: The findings of 2 distinct gene clusters in EA and different biological pathways in each gene cluster suggest 2 different pathogenesis mechanisms underlying EA.


Subject(s)
Aged , Airway Obstruction , Asthma , Cluster Analysis , Dataset , Eosinophils , Epithelial-Mesenchymal Transition , Gene Expression , Humans , Multigene Family , Sputum , Transcriptome
17.
Article in English | WPRIM | ID: wpr-719501

ABSTRACT

BACKGROUND: As of 2011, among 250 administrative districts in Korea, 54 districts did not have obstetrics and gynecology clinics or hospitals providing prenatal care and delivery services. The Korean government designated 38 regions among 54 districts as “Obstetric Care Underserved Areas (OCUA).” However, little is known there are any differences in pregnancy, prenatal care, and outcomes of women dwelling in OCUA compared to women in other areas. The purposes of this study were to compare the pregnancy related indicators (PRIs) and adequacy of prenatal care between OCUA region and non-OCUA region. METHODS: Using National Health Insurance database in Korea from January 1, 2012 to December 31, 2014, we constructed the whole dataset of women who terminated pregnancy including delivery and abortion. We assessed incidence rate of 17 PRIs and adequacy of prenatal care. All indicators were compared between OCUA group and non-OCUA group. RESULTS: The women dwelling in OCUA regions were more likely to get abortion (4.6% in OCUA vs. 3.6% in non-OCUA) and receive inadequate prenatal care (7.2% vs. 4.4%). Regarding abortion rate, there were significant regional differences in abortion rate. The highest abortion rate was 10.3% and the lowest region was 1.2%. Among 38 OCUA regions, 29 regions' abortion rates were higher than the national average of abortion rate (3.56%) and there were 10 regions in which abortion rates were higher than 7.0%. In addition, some PRIs such as acute pyelonephritis and transfusion in obstetric hemorrhage were more worse in OCUA regions compared to non-OCUA regions. CONCLUSION: PRIs are different according to the regions where women are living. The Korean government should make an effort reducing these gaps of obstetric cares between OCUA and non-OCUA.


Subject(s)
Abortion, Induced , Dataset , Female , Gynecology , Hemorrhage , Humans , Incidence , Korea , Medically Underserved Area , National Health Programs , Obstetrics , Pregnancy , Pregnant Women , Prenatal Care , Pyelonephritis
18.
Article in Korean | WPRIM | ID: wpr-761395

ABSTRACT

OBJECTIVE: Myeloproliferative neoplasm (MPN) is considered as one of the risk factors of ischemic stroke. Some MPN patients manifest stroke as their first symptom. Our purpose was to assess diagnostic rate of MPN in newly diagnosed acute ischemic stroke patients. METHODS: This study was performed using National Health Insurance Service Ilsan Hospital dataset. Data retrieving was performed by defining by defining the patient with coding of acute ischemic stroke from January 2013 to June 2017. We selected only the patients who had checked brain magnetic resonance imaging and complete blood cell count (CBC) in emergency room or on admission. Among the results of CBC finding, hemoglobin and platelet count were analyzed. Erythrocytosis was defined >16.5 g/dL (male), >16 g/dL (female) according to revised World Health Organization (WHO) classification of polycythemia vera (PV) criteria. Thrombocytosis was >450,000/µL according to revised WHO classification of essential thrombocythemia (ET). RESULTS: Total number of newly diagnosed acute ischemic stroke was 1,613 patients. Seven patients (0.43%) were diagnosed MPN (ET=2, PV=5) after ischemic stroke. Patients who had thrombocytosis and erythrocytosis were 18 and 105, respectively. Three patients who had thrombocytosis were diagnosed MPN (ET=2, PV=1). Two patients with erythrocytosis were diagnosed MPN (PV=2). Two patients had both thrombocytosis and erythrocytosis, and two of them were diagnosed PV. Seventy-one patients who had erythrocytosis were normalized in follow-up period. Six patients who had thrombocytosis and 30 patients who had erythrocytosis did not further evaluate. CONCLUSION: CBC has to be carefully read and MPN can be suspected. Diagnosis must be confirmed by hematologist to initiate appropriate treatment. It is important to recognized suspected MPN patients to prevent stroke.


Subject(s)
Blood Cell Count , Brain , Classification , Clinical Coding , Dataset , Diagnosis , Emergency Service, Hospital , Follow-Up Studies , Humans , Magnetic Resonance Imaging , National Health Programs , Platelet Count , Polycythemia , Polycythemia Vera , Risk Factors , Stroke , Thrombocythemia, Essential , Thrombocytosis , World Health Organization
19.
Article in English | WPRIM | ID: wpr-765385

ABSTRACT

OBJECTIVE: Since less invasive endovascular treatment was introduced to South Korea in 1994, a considerable proportion of endovascular treatments have been performed by neuroradiology doctors, and endovascular treatments by vascular neurosurgeons have recently increased. However, few specific statistics are known regarding how many endovascular treatments are performed by neurosurgeons. Thus, authors compared endovascular treatments collaboratively performed by vascular neurosurgeons with all cases throughout South Korea from 2013 to 2017 to elucidate the role of neurosurgeons in the field of endovascular treatment in South Korea. METHODS: The Society of Korean Endovascular Neurosurgeons (SKEN) has issued annual reports every year since 2014. These reports cover statistics on endovascular treatments collaboratively or individually performed by SKEN members from 2013 to 2017. The data was requested and collected from vascular neurosurgeons in various hospitals. The study involved 77 hospitals in its first year, and 100 in its last. National statistics on endovascular treatment from all over South Korea were obtained from the Healthcare Bigdata Hub website of the Health Insurance Review & Assessment Service based on the Electronic Data Interchange (EDI) codes (in the case of intra-arterial (IA) thrombolysis, however, statistics were based on a combination of the EDI and I63 codes, a cerebral infarction disease code) from 2013 to 2017. These two data sets were directly compared and the ratios were obtained. RESULTS: Regionally, during the entire study period, endovascular treatments by SKEN members were most common in Gyeonggi-do, followed by Seoul and Busan. Among the endovascular treatments, conventional cerebral angiography was the most common, followed by cerebral aneurysmal coiling, endovascular treatments for ischemic stroke, and finally endovascular treatments for vascular malformation and tumor embolization. The number of endovascular treatments performed by SKEN members increased every year. CONCLUSION: The SKEN members have been responsible for the major role of endovascular treatments in South Korea for the recent 5 years. This was achieved through the perseverance of senior members who started out in the midst of hardship, the establishment of standards for the training/certification of endovascular neurosurgery, and the enthusiasm of current SKEN members who followed. To provide better treatment to patients, we will have to make further progress in SKEN.


Subject(s)
Cerebral Angiography , Cerebral Infarction , Data Interpretation, Statistical , Dataset , Delivery of Health Care , Endovascular Procedures , Humans , Insurance, Health , Intracranial Aneurysm , Korea , Neurosurgeons , Neurosurgery , Seoul , Stroke , Vascular Malformations
20.
Article in English | WPRIM | ID: wpr-765362

ABSTRACT

OBJECTIVE: Bone mineral density (BMD) is an important consideration during fusion surgery. Although dual X-ray absorptiometry is considered as the gold standard for assessing BMD, quantitative computed tomography (QCT) provides more accurate data in spine osteoporosis. However, QCT has the disadvantage of additional radiation hazard and cost. The present study was to demonstrate the utility of artificial intelligence and machine learning algorithm for assessing osteoporosis using Hounsfield units (HU) of preoperative lumbar CT coupling with data of QCT. METHODS: We reviewed 70 patients undergoing both QCT and conventional lumbar CT for spine surgery. The T-scores of 198 lumbar vertebra was assessed in QCT and the HU of vertebral body at the same level were measured in conventional CT by the picture archiving and communication system (PACS) system. A multiple regression algorithm was applied to predict the T-score using three independent variables (age, sex, and HU of vertebral body on conventional CT) coupling with T-score of QCT. Next, a logistic regression algorithm was applied to predict osteoporotic or non-osteoporotic vertebra. The Tensor flow and Python were used as the machine learning tools. The Tensor flow user interface developed in our institute was used for easy code generation. RESULTS: The predictive model with multiple regression algorithm estimated similar T-scores with data of QCT. HU demonstrates the similar results as QCT without the discordance in only one non-osteoporotic vertebra that indicated osteoporosis. From the training set, the predictive model classified the lumbar vertebra into two groups (osteoporotic vs. non-osteoporotic spine) with 88.0% accuracy. In a test set of 40 vertebrae, classification accuracy was 92.5% when the learning rate was 0.0001 (precision, 0.939; recall, 0.969; F1 score, 0.954; area under the curve, 0.900). CONCLUSION: This study is a simple machine learning model applicable in the spine research field. The machine learning model can predict the T-score and osteoporotic vertebrae solely by measuring the HU of conventional CT, and this would help spine surgeons not to under-estimate the osteoporotic spine preoperatively. If applied to a bigger data set, we believe the predictive accuracy of our model will further increase. We propose that machine learning is an important modality of the medical research field.


Subject(s)
Absorptiometry, Photon , Artificial Intelligence , Boidae , Bone Density , Classification , Dataset , Humans , Learning , Logistic Models , Machine Learning , Osteoporosis , Spine , Surgeons
SELECTION OF CITATIONS
SEARCH DETAIL