• 제목/요약/키워드: Radiography training

검색결과 50건 처리시간 0.031초

Convolutional neural network of age-related trends digital radiographs of medial clavicle in a Thai population: a preliminary study

  • Phisamon Kengkard;Jirachaya Choovuthayakorn;Chollada Mahakkanukrauh;Nadee Chitapanarux;Pittayarat Intasuwan;Yanumart Malatong;Apichat Sinthubua;Patison Palee;Sakarat Na Lampang;Pasuk Mahakkanukrauh
    • Anatomy and Cell Biology
    • /
    • 제56권1호
    • /
    • pp.86-93
    • /
    • 2023
  • Age at death estimation has always been a crucial yet challenging part of identification process in forensic field. The use of human skeletons have long been explored using the principle of macro and micro-architecture change in correlation with increasing age. The clavicle is recommended as the best candidate for accurate age estimation because of its accessibility, time to maturation and minimal effect from weight. Our study applies pre-trained convolutional neural network in order to achieve the most accurate and cost effective age estimation model using clavicular bone. The total of 988 clavicles of Thai population with known age and sex were radiographed using Kodak 9000 Extra-oral Imaging System. The radiographs then went through preprocessing protocol which include region of interest selection and quality assessment. Additional samples were generated using generative adversarial network. The total clavicular images used in this study were 3,999 which were then separated into training and test set, and the test set were subsequently categorized into 7 age groups. GoogLeNet was modified at two layers and fine tuned the parameters. The highest validation accuracy was 89.02% but the test set achieved only 30% accuracy. Our results show that the use of medial clavicular radiographs has a potential in the field of age at death estimation, thus, further study is recommended.

A Comparative Study of Deep Learning Techniques for Alzheimer's disease Detection in Medical Radiography

  • Amal Alshahrani;Jenan Mustafa;Manar Almatrafi;Layan Albaqami;Raneem Aljabri;Shahad Almuntashri
    • International Journal of Computer Science & Network Security
    • /
    • 제24권5호
    • /
    • pp.53-63
    • /
    • 2024
  • Alzheimer's disease is a brain disorder that worsens over time and affects millions of people around the world. It leads to a gradual deterioration in memory, thinking ability, and behavioral and social skills until the person loses his ability to adapt to society. Technological progress in medical imaging and the use of artificial intelligence, has provided the possibility of detecting Alzheimer's disease through medical images such as magnetic resonance imaging (MRI). However, Deep learning algorithms, especially convolutional neural networks (CNNs), have shown great success in analyzing medical images for disease diagnosis and classification. Where CNNs can recognize patterns and objects from images, which makes them ideally suited for this study. In this paper, we proposed to compare the performances of Alzheimer's disease detection by using two deep learning methods: You Only Look Once (YOLO), a CNN-enabled object recognition algorithm, and Visual Geometry Group (VGG16) which is a type of deep convolutional neural network primarily used for image classification. We will compare our results using these modern models Instead of using CNN only like the previous research. In addition, the results showed different levels of accuracy for the various versions of YOLO and the VGG16 model. YOLO v5 reached 56.4% accuracy at 50 epochs and 61.5% accuracy at 100 epochs. YOLO v8, which is for classification, reached 84% accuracy overall at 100 epochs. YOLO v9, which is for object detection overall accuracy of 84.6%. The VGG16 model reached 99% accuracy for training after 25 epochs but only 78% accuracy for testing. Hence, the best model overall is YOLO v9, with the highest overall accuracy of 86.1%.

Diagnostic Performance of a New Convolutional Neural Network Algorithm for Detecting Developmental Dysplasia of the Hip on Anteroposterior Radiographs

  • Hyoung Suk Park;Kiwan Jeon;Yeon Jin Cho;Se Woo Kim;Seul Bi Lee;Gayoung Choi;Seunghyun Lee;Young Hun Choi;Jung-Eun Cheon;Woo Sun Kim;Young Jin Ryu;Jae-Yeon Hwang
    • Korean Journal of Radiology
    • /
    • 제22권4호
    • /
    • pp.612-623
    • /
    • 2021
  • Objective: To evaluate the diagnostic performance of a deep learning algorithm for the automated detection of developmental dysplasia of the hip (DDH) on anteroposterior (AP) radiographs. Materials and Methods: Of 2601 hip AP radiographs, 5076 cropped unilateral hip joint images were used to construct a dataset that was further divided into training (80%), validation (10%), or test sets (10%). Three radiologists were asked to label the hip images as normal or DDH. To investigate the diagnostic performance of the deep learning algorithm, we calculated the receiver operating characteristics (ROC), precision-recall curve (PRC) plots, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) and compared them with the performance of radiologists with different levels of experience. Results: The area under the ROC plot generated by the deep learning algorithm and radiologists was 0.988 and 0.988-0.919, respectively. The area under the PRC plot generated by the deep learning algorithm and radiologists was 0.973 and 0.618-0.958, respectively. The sensitivity, specificity, PPV, and NPV of the proposed deep learning algorithm were 98.0, 98.1, 84.5, and 99.8%, respectively. There was no significant difference in the diagnosis of DDH by the algorithm and the radiologist with experience in pediatric radiology (p = 0.180). However, the proposed model showed higher sensitivity, specificity, and PPV, compared to the radiologist without experience in pediatric radiology (p < 0.001). Conclusion: The proposed deep learning algorithm provided an accurate diagnosis of DDH on hip radiographs, which was comparable to the diagnosis by an experienced radiologist.

치과방사선 질관리 향상을 위한 교육자 대비 비교육자 비교연구 - 치과방사선학 이론 및 실습교육과 임상실습교육을 중심으로 - (A comparative study of educators vs, non-educators designed to improve dental radiographic quality control - Focusing on theories of dental radiographic and practical training and clinical practice education -)

  • 김승희;홍수민;이광옥
    • 한국방사선학회논문지
    • /
    • 제6권5호
    • /
    • pp.421-426
    • /
    • 2012
  • 본 연구의 목적은 치위생 전공 학생들의 치과방사선 장비 및 물품의 질관리에 관한 지식수준을 파악하고, 방사선 질관리에 대한 이론 및 실습 교육정도를 조사하여 치과위생사 양성과정에서의 체계적인 방사선 질관리와 관련된 교육과정 개설 및 개편을 위해 필요한 기초자료를 제공하고자 한다. 연구의 목적을 달성하기 위해서 치위생 전공 학생 중 치과방사선 과목을 수강한 학생 453명을 대상으로 설문조사를 실시하였다. 분석 가능한 자료를 SPSS 12.0을 활용하여 자료를 분석하였으며, 연구대상자의 변인별 특성을 알아보기 위해 빈도분석, 신뢰도분석, 카이제곱 검정, 독립 T-test, 일원배치분산분석 후 사후검정으로 scheffe 방법을 실시하였다. 분석결과, 첫째, 방사선 질관리에 대한 지식수준은 12점 만점 중 평균 $7.71{\pm}1.7$점으로 나타났으며 치과방사선 교과목 이수 시 이론수업과 실습수업을 받을수록, 지식수준이 높게 나타났다(p<0.001). 둘째, 방사선 질관리에 대한 임상실습교육 수준은 13개 항목 중 1~3개를 경험한 학생수가 가장 많은 것으로 나타났으며, 임상실습교육을 전혀 받지 않은 학생도 26.3%로 조사되어 방사선 질관리에 관한 적절한 실습교육을 위탁교육기관에서 제공해야 할 필요성이 있었다. 셋째, 방사선 질관리에 대한 실습교육 13개 항목중 질관리 실습을 전혀 경험하지 못한 사람의 정답 문항 수는 평균 7.20개, 1~3개 항목을 교육받은 사람의 정답 문항수는 평균 7.84개, 4~5개 항목을 교육받은 사람의 정답 문항수는 평균 7.87개, 6개 이상 항목을 교육받은 사람의 정답 문항수는 8.14개로 나타났으며, 임상실습교육기간 중 질관리 관련 교육 경험수가 많을수록 지식수준이 높은 것으로 나타났다.

치과위생사의 방사선 안전관리에 대한 조사 연구 (A Study on Radiation Safety Management by Dental Hygienist)

  • 강은주;이경희;김영임
    • 치위생과학회지
    • /
    • 제5권3호
    • /
    • pp.105-112
    • /
    • 2005
  • 구강진료기관에서 이용되는 방사선은 노출양이 극히 미량이라고 알려져 있지만, 장기간 방사선을 취급하는 경우에는 위해작용이 나타날 수 있으므로 이에 대한 방사선 종사자의 인식변화가 필요하다고 볼 수 있다. 따라서, 본 연구에서는 치과위생사의 방사선 안전관리에 대한 지식, 태도 및 행위를 파악하여 구강 방사선 촬영실에서의 안전관리 행위에 영향을 미치는 요인을 분석함으로써 치과위생사뿐만 아니라 일반 이용자들의 방사선에 의한 건강장해를 예방하기 위한 방어계획 수립과 치과위생사의 방사선 안전에 대한 보건교육 프로그램을 설계함에 도움이 되는시사점을 도출하고자 각 지역의 치과병원 및 치과의원에 근무하는 치과위생사를 대상으로 2003년 12월부터 2004년 3월까지 약 4개월동안 조사하여 다음과 같은 결과를 얻었다. 1. 방사선 안전관리에 대한 지식 수준을 살펴본 결과 총15점 만점 중 평균이 $8.59{\pm}2.36$점으로 나타났으며, 최고점수는 13점, 최소점수는 3점으로 나타났다. 또한 일반적인 특성에 따른 지식 수준을 살펴보면, 근무경력별로(p < 0.001),결혼 여부별로(p < 0.001), 방사선 안전교육 여부별로(p < 0.001), 병원형태별로(p < 0.001) 통계적으로 유의한 차이를 보였다. 2. 방사선 안전관리에 대한 태도 수준을 살펴본 결과, 5점 만점 중 전체 평균이 $4.08{\pm}0.50$점으로 나타났으며, 문항별 최고점수는 평균 $4.31{\pm}0.73$점, 최저 점수는$3.82{\pm}0.89$점으로 나타났다. 또한 일반적인 특성에 따른 태도 수준을 살펴보면, 연령별로(p < 0.001), 근무경력별로(p < 0.05), 방사선 안전교육 여부별로(p < 0.01), 병원형태별로(p < 0.001) 통계적으로 유의한 차이를 보였다. 3. 방사선 안전관리에 대한 행위 수준을 살펴본 결과, 5점 만점 중 전체평균은 $2.89{\pm}0.77$점으로 태도 수준에 비해 행위 수준은 낮게 조사되었으며, 문항별 최고점수는 $3.82{\pm}0.94$점, 최저점수는 $2.37{\pm}1.04$점으로 나타났다. 또한 일반적인 특성에 따른 행위수준을 살펴본 결과, 근무경력별로(p < 0.001), 병원형태별로(p < 0.001) 통계적으로 유의한 차이를 보였다. 4. 방사선 안전관리 지식, 태도 및 행위와의 관계를 살펴본 결과 방사선 안전관리에 대한 지식이 높을수록 방사선 안전관리 태도와 행위 정도가 높았으며, 방사선 안전관리에 대한 태도가 높을수록 방사선 안전관리 행위의 정도가 높은 것으로 나타났다.

  • PDF

방사선사(放射線士) 교육(敎育)의 임상실습(臨床實習) 개선(改善)에 관(關)한 연구(硏究) (A Study of the Improvement of Clinical and Practical Trainings in the Education of Radiologic Technologists)

  • 이만구;강세식;윤한식;허준
    • 대한방사선기술학회지:방사선기술과학
    • /
    • 제6권1호
    • /
    • pp.117-129
    • /
    • 1983
  • This study, in order to improve clinical and practical trainings in the education of radiologic technologists, applies to 76 medical institutions of 91 ones which are used as the hospitals of clinical and practical training in 9 existing junior colleges except 3 new ones of 12 ones throughout all over the country from November 1, in 1982 to April 30, in 1983. And the purpose of this study is to research the percent conditions of basic practical trainings and clinical ones enforced in each college, and the percent conditions, equipments, contents, and opinions in clinical and practical trainings enforced in each hospital. The results are summarized as follows; 1. In the case of junior colleges in the whole country the curriculum of basic practical trainings averages 336.66 hours and the limits are between 120 and 510 hours. The actual hours in practice average 140 hours and the limits are between 60 and 240 hours, which correspond to 41.58% of the curriculum of basic practical trainings. 2. There were three junior colleges among nine that had a reserved hospital for clinical and practical trainings(only 33.33%). 3. The period of the practice was almost vacation in 4 junior colleges. The practice was conducted only for students to want the practice(44.45%), junior colleges that all students in them conducted the practice was 2 junior colleges and presented 22.22%. 4. In the field of students engaging in the practice, each field of radiation therapy and nuclear medicine presented 16.5%, 20.3% and almost students didin't have experience for the practice. 5. In medical institutions the educational institutions for intern showed 67.11%. Hospital with radiologist showed 26.32%. Radiotechnologist who had experience below 5 years presented 60.17%. 6. In the equipment for radiation diagnosis, each hospital had no difference. The number of hospitals passessing diagnostic equipments above 125 KVP was 56.26%. But radiation therapy equipment and nuclear medicine equipment had extremely low rate. 7. In the diagnosis of patient in the practice hospital, conventional radiography-to Skull, Chest, Abdomen, Skeleton, Urogenital system-reached the criterion. But special radiography was comparatively low. There appeared low rate, 32.89% in the field of nuclear medicine, 15.79% in the field of radiation therapy. 8. Students who carried out the practice were 1-89 students, days in practice were 1-30 days. There were differences in that point among among hospitals. Junior colleges conducting the practice were 2 colleges per hospital. Scope of the object were 1-8 junior colleges. 9. The practice conducted for the request of the colleges presented 72.37%, in addition, The prctices were conducted for growth of the younger generation and the same coperation with the colleges establishment of sisterhood with the colleges, relationship with students. 10. The practice conducted without the establishment of plan presented 59.21% The need for guiding book to the practice and evaluating was recognized over 90%. 11. In the relation between the practice with achievement of credit. There were big differences in opinion between hospitals-Group and the colleges-Group; hospital-Group had opinion that must follow achievement of credit with the practice. The colleges-Group had opinion that must conduct the practice after achieving credit. 12. After conducting the practice, in the practice leaders satisfaction degree dissatisfactory opinion presented the most rate 80.26%. Very much satisfactory opinion, as one hospital, presentd only 1.32%. 13. Both hospitals-Group and the colleges-Group had an opinion that the practice leader must have actual experiences, lectures and achievement, an opinion that actual experiences is over 5 years. 14. In the guide of human relation, cooperation, responsibility, courtesy to patients. Both hospitals-Group and the colleges-Group had an opinion that the guide must be involved in the period of the practice and must be instructed.

  • PDF

치주질환 예측을 위한 치과 X-선 영상에서의 초해상화 알고리즘 적용 가능성 연구 (Investigation of the Super-resolution Algorithm for the Prediction of Periodontal Disease in Dental X-ray Radiography)

  • 김한나
    • 한국방사선학회논문지
    • /
    • 제15권2호
    • /
    • pp.153-158
    • /
    • 2021
  • 치주질환의 조기 진단률 및 예측 정확도 향상을 위한 X-선 영상 분석은 매우 중요한 분야이다. 이러한 치과 X-선 영상의 화질 개선을 위한 인공 지능 기반의 알고리즘 개발 및 적용에 관한 연구는 전 세계적으로 널리 수행 중이다. 따라서 본 연구의 목표는 치주질환 예측을 위한 치과 X-선 영상에서의 초해상화 알고리즘의 모델링 및 적용 가능성에 관하여 평가하는 것이다. 초해상화 알고리즘은 convolution layer와 ReLU를 기반으로 구성하였고, 저해상도 영상을 2배로 업샘플링 한 영상을 입력으로 사용하였다. 딥러닝 훈련을 위해 사용한 치과 X-선 데이터는 1,500장을 사용하였다. 영상의 정량적 평가는 2가지 영상의 비교를 통해 유사도를 측정할 수 있는 인자인 root mean square error와 structural similarity를 사용하였다. 이와 더불어 최근에 개발된 no-reference 기반으로 사용되는 natural image quality evaluator 와 blind/referenceless image spatial quality evaluator를 추가적으로 분석하였다. 결과적으로 기존에 사용되던 bicubic 기반의 업샘플링 기법을 사용하였을 때에 비하여 제안하는 방법이 치과 X-선 영상에서 평균적으로 유사도와 no-reference 기반의 평가 인자가 각각 1.86 그리고 2.14배 향상됨을 확인하였다. 결론적으로 치주질환의 예측을 위한 초해상화 알고리즘의 치과 X-선 영상에서의 유용성을 증명하였고 향후 다양한 분야에서의 적용 가능성이 높을 것으로 기대된다.

준지도학습 방법을 이용한 흉부 X선 사진에서 척추측만증의 진단 (Diagnosis of Scoliosis Using Chest Radiographs with a Semi-Supervised Generative Adversarial Network)

  • 이우진;신기원;이준수;유승진;윤민아;최요원;홍길선;김남국;백상현
    • 대한영상의학회지
    • /
    • 제83권6호
    • /
    • pp.1298-1311
    • /
    • 2022
  • 목적 흉부 X선 사진에서 척추측만증을 조기진단 할 수 있는 딥러닝 기반의 스크리닝 소프트웨어를 준지도학습(semi-supervised generative adversarial network; 이하 GAN) 방법을 이용하여 개발하고자 하였다. 대상과 방법 두 곳의 상급종합병원에서 촬영된 흉부 X선 사진에서 척추측만증을 조기진단할 수 있는 스크리닝 소프트웨어를 개발하기 위하여 GAN 방법이 이용되었다. GAN의 훈련과정에서 경증에서 중증의 척추측만증을 보이는 흉부 X선 사진들을 사용하였으며 upstream task에서 척추측만증의 특징을 학습하고, downstream task에서 정상과 척추측만증을 분류하도록 훈련하였다. 결과 수신자 조작 특성 곡선의 곡선하면적(area under the receiver operating characteristic curve), 음성예측도, 양성예측도, 민감도 및 특이도는 각각 0.856, 0.950, 0.579, 0.985, 0.285이었다. 결론 우리가 GAN 방법을 이용하여 개발한 딥러닝 기반의 스크리닝 소프트웨어는 청소년의 흉부 X선에서 척추측만증을 진단하는데 있어서 높은 음성예측도와 민감도를 보였다. 이 소프트웨어가 건강검진을 목적으로 촬영한 청소년의 흉부 X선 사진에 진단 스크리닝 도구로써 이용된다면 영상의학과 의사의 부담을 덜어주며, 척추측만증의 조기진단에 기여할 것으로 생각된다.

Deep Learning-Assisted Diagnosis of Pediatric Skull Fractures on Plain Radiographs

  • Jae Won Choi;Yeon Jin Cho;Ji Young Ha;Yun Young Lee;Seok Young Koh;June Young Seo;Young Hun Choi;Jung-Eun Cheon;Ji Hoon Phi;Injoon Kim;Jaekwang Yang;Woo Sun Kim
    • Korean Journal of Radiology
    • /
    • 제23권3호
    • /
    • pp.343-354
    • /
    • 2022
  • Objective: To develop and evaluate a deep learning-based artificial intelligence (AI) model for detecting skull fractures on plain radiographs in children. Materials and Methods: This retrospective multi-center study consisted of a development dataset acquired from two hospitals (n = 149 and 264) and an external test set (n = 95) from a third hospital. Datasets included children with head trauma who underwent both skull radiography and cranial computed tomography (CT). The development dataset was split into training, tuning, and internal test sets in a ratio of 7:1:2. The reference standard for skull fracture was cranial CT. Two radiology residents, a pediatric radiologist, and two emergency physicians participated in a two-session observer study on an external test set with and without AI assistance. We obtained the area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity along with their 95% confidence intervals (CIs). Results: The AI model showed an AUROC of 0.922 (95% CI, 0.842-0.969) in the internal test set and 0.870 (95% CI, 0.785-0.930) in the external test set. The model had a sensitivity of 81.1% (95% CI, 64.8%-92.0%) and specificity of 91.3% (95% CI, 79.2%-97.6%) for the internal test set and 78.9% (95% CI, 54.4%-93.9%) and 88.2% (95% CI, 78.7%-94.4%), respectively, for the external test set. With the model's assistance, significant AUROC improvement was observed in radiology residents (pooled results) and emergency physicians (pooled results) with the difference from reading without AI assistance of 0.094 (95% CI, 0.020-0.168; p = 0.012) and 0.069 (95% CI, 0.002-0.136; p = 0.043), respectively, but not in the pediatric radiologist with the difference of 0.008 (95% CI, -0.074-0.090; p = 0.850). Conclusion: A deep learning-based AI model improved the performance of inexperienced radiologists and emergency physicians in diagnosing pediatric skull fractures on plain radiographs.

Deep learning-based automatic segmentation of the mandibular canal on panoramic radiographs: A multi-device study

  • Moe Thu Zar Aung;Sang-Heon Lim;Jiyong Han;Su Yang;Ju-Hee Kang;Jo-Eun Kim;Kyung-Hoe Huh;Won-Jin Yi;Min-Suk Heo;Sam-Sun Lee
    • Imaging Science in Dentistry
    • /
    • 제54권1호
    • /
    • pp.81-91
    • /
    • 2024
  • Purpose: The objective of this study was to propose a deep-learning model for the detection of the mandibular canal on dental panoramic radiographs. Materials and Methods: A total of 2,100 panoramic radiographs (PANs) were collected from 3 different machines: RAYSCAN Alpha (n=700, PAN A), OP-100 (n=700, PAN B), and CS8100 (n=700, PAN C). Initially, an oral and maxillofacial radiologist coarsely annotated the mandibular canals. For deep learning analysis, convolutional neural networks (CNNs) utilizing U-Net architecture were employed for automated canal segmentation. Seven independent networks were trained using training sets representing all possible combinations of the 3 groups. These networks were then assessed using a hold-out test dataset. Results: Among the 7 networks evaluated, the network trained with all 3 available groups achieved an average precision of 90.6%, a recall of 87.4%, and a Dice similarity coefficient (DSC) of 88.9%. The 3 networks trained using each of the 3 possible 2-group combinations also demonstrated reliable performance for mandibular canal segmentation, as follows: 1) PAN A and B exhibited a mean DSC of 87.9%, 2) PAN A and C displayed a mean DSC of 87.8%, and 3) PAN B and C demonstrated a mean DSC of 88.4%. Conclusion: This multi-device study indicated that the examined CNN-based deep learning approach can achieve excellent canal segmentation performance, with a DSC exceeding 88%. Furthermore, the study highlighted the importance of considering the characteristics of panoramic radiographs when developing a robust deep-learning network, rather than depending solely on the size of the dataset.