Pub Date : 2018-03-01DOI: 10.7599/HMR.2018.38.1.56
Y. Jung, D. Choi
Patients undergoing hepatobiliary and pancreatic (HBP) surgery often need to be transfused, despite advances in surgical skills and perioperative care. However, many studies have indicated that cancer patients who are transfused have higher rates of perioperative mortality and cancer recurrence, and poorer prognoses [1]. Moreover, viral or bacterial infections, immunologic reactions, and increased postoperative morbidity are other adverse consequences of allogeneic transfusions. Furthermore, since there are not enough blood donors in Korea to supply the demand, new treatment strategies for HBP patients are needed. Patient blood management (PBM) programs, medical care without allogeneic blood transfusion, have traditionally been applied in various clinical situations, e.g., when patients refuse to be transfused for religious reasons, when there is no blood to transfuse, and when safe blood is not available [2]. Although PBM is a relatively new technology in the field of HBP surgery, its general concepts are very similar to those of traditional PBM. The basic concepts of PBM applicable to the perioperative and intraoperative method have recently been described. Erythropoietin, ferritin, vitamin B12, or volume expanders and preoperative autologous blood donation (PAD) are used in perioperative PBM. Intraoperative management includes acute normovolemic hemodilution (ANH), cell salvage (Cell Saver®), and hypotensive anesthesia. Although the disadvantages of transfusion and the advantages of PBM are widely recognized, few studies have evaluated the beneficial effects of PBM in HBP surgery. Although the use of PBM in HBP operations without transfusion (including pancreaticoduodenectomy for periampullary lesions, living donor liver transplantation, and major hepatectomy) has been reported in the past few years, it is inherently challenging to carry out researches on transfusion-related issues because reasons and sequelae of transfusion are multifactorial [3-6]. The goal of this article is to review the current status of PBM programs in HBP surgery. Review
{"title":"Patient Blood Management in Hepatobiliary and Pancreatic Surgery","authors":"Y. Jung, D. Choi","doi":"10.7599/HMR.2018.38.1.56","DOIUrl":"https://doi.org/10.7599/HMR.2018.38.1.56","url":null,"abstract":"Patients undergoing hepatobiliary and pancreatic (HBP) surgery often need to be transfused, despite advances in surgical skills and perioperative care. However, many studies have indicated that cancer patients who are transfused have higher rates of perioperative mortality and cancer recurrence, and poorer prognoses [1]. Moreover, viral or bacterial infections, immunologic reactions, and increased postoperative morbidity are other adverse consequences of allogeneic transfusions. Furthermore, since there are not enough blood donors in Korea to supply the demand, new treatment strategies for HBP patients are needed. Patient blood management (PBM) programs, medical care without allogeneic blood transfusion, have traditionally been applied in various clinical situations, e.g., when patients refuse to be transfused for religious reasons, when there is no blood to transfuse, and when safe blood is not available [2]. Although PBM is a relatively new technology in the field of HBP surgery, its general concepts are very similar to those of traditional PBM. The basic concepts of PBM applicable to the perioperative and intraoperative method have recently been described. Erythropoietin, ferritin, vitamin B12, or volume expanders and preoperative autologous blood donation (PAD) are used in perioperative PBM. Intraoperative management includes acute normovolemic hemodilution (ANH), cell salvage (Cell Saver®), and hypotensive anesthesia. Although the disadvantages of transfusion and the advantages of PBM are widely recognized, few studies have evaluated the beneficial effects of PBM in HBP surgery. Although the use of PBM in HBP operations without transfusion (including pancreaticoduodenectomy for periampullary lesions, living donor liver transplantation, and major hepatectomy) has been reported in the past few years, it is inherently challenging to carry out researches on transfusion-related issues because reasons and sequelae of transfusion are multifactorial [3-6]. The goal of this article is to review the current status of PBM programs in HBP surgery. Review","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121013152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-03-01DOI: 10.7599/HMR.2018.38.1.67
T. Um
Blood transfusion is an essential medical procedure that can save the patient’s life. But, it is anticipated that blood transfusion products will be lacking in Korea in the near future. This is due to the fact that eligible blood donors—the young population—are decreasing, whereas blood recipients—the elderly population—are increasing. Low birth rate and aging society have become big social problems in Korea recently. Korea’s birth rate is the lowest among OECD countries, which is 1.17 in 2016 [1]. The elderly population aged 65 or older is 13.8% in 2017 and it is expected to be over 20% in 2026, becoming a super-aged society. Aging populations present higher risks of malignancies and chronic diseases; and are more likely to require complex surgical interventions [2]. If unnecessary blood transfusions are to be decreased, we would be able to prevent waste of precious blood resources and to save significant amount of healthcare costs [3]. In Australia, the NBA estimated that a 5% reduction in RBC use would result in a national saving of AUD14.6 million [4]. Beyond the economic savings, this also means ameliorating blood transfusion related risks to the patients. Blood transfusion is still not free of the risks of complications such as infection and immunomodulation, although they are dramatically decreased through the advances in transfusion medicine. Furthermore, this is providing the best care to the patients because it is now well known that transfusion may lead to poorer patient outcomes, such as survival rates [5-7]. So, increasing the adequacy of blood transfusion is the strategy for not only preventing wastage of precious blood resources and blood shortage, but also providing patients with the best treatments by decreasing risk of complications.
{"title":"Patient Blood Management: Future Perspective in Korea","authors":"T. Um","doi":"10.7599/HMR.2018.38.1.67","DOIUrl":"https://doi.org/10.7599/HMR.2018.38.1.67","url":null,"abstract":"Blood transfusion is an essential medical procedure that can save the patient’s life. But, it is anticipated that blood transfusion products will be lacking in Korea in the near future. This is due to the fact that eligible blood donors—the young population—are decreasing, whereas blood recipients—the elderly population—are increasing. Low birth rate and aging society have become big social problems in Korea recently. Korea’s birth rate is the lowest among OECD countries, which is 1.17 in 2016 [1]. The elderly population aged 65 or older is 13.8% in 2017 and it is expected to be over 20% in 2026, becoming a super-aged society. Aging populations present higher risks of malignancies and chronic diseases; and are more likely to require complex surgical interventions [2]. If unnecessary blood transfusions are to be decreased, we would be able to prevent waste of precious blood resources and to save significant amount of healthcare costs [3]. In Australia, the NBA estimated that a 5% reduction in RBC use would result in a national saving of AUD14.6 million [4]. Beyond the economic savings, this also means ameliorating blood transfusion related risks to the patients. Blood transfusion is still not free of the risks of complications such as infection and immunomodulation, although they are dramatically decreased through the advances in transfusion medicine. Furthermore, this is providing the best care to the patients because it is now well known that transfusion may lead to poorer patient outcomes, such as survival rates [5-7]. So, increasing the adequacy of blood transfusion is the strategy for not only preventing wastage of precious blood resources and blood shortage, but also providing patients with the best treatments by decreasing risk of complications.","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126553431","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-03-01DOI: 10.7599/HMR.2018.38.1.49
Taehee Kim, M. Jeong
Blood transfusion is generally considered to be the solution of anemia and blood loss during surgery. Transfusion is a very efficient and effective method to correct anemia, but there has been increasing evidence that blood transfusion does not lead to improved outcomes and that morbidity and mortality increase in a dose-dependent manner [1,2]. It has been shown that even a single unit of transfused packed red blood cells (PRBCs) can increase 30day mortality, complicated mortality, pneumonia and sepsis [3]. Therefore, it is preferable to avoid unnecessary blood transfusion or to minimize blood transfusion. In surgical patients, patient blood management focuses on anemia management, minimization of blood loss, appropriate transfusion for reducing surgical risk, and improving patient outcome after surgery. Recognition and management of pre-operative anemia represent an opportunity to optimize patient status before surgery, thereby reducing blood transfusion and potentially improving recovery from surgery and associated postoperative outcomes. A complex approach such as anesthetic strategy and operative techniques, pharmacological intervention, and cell salvage is required to reduce bleeding during surgery. In this review, we reviewed the studies about blood management in the stance of anesthesiologists. Management of coagulopathy and blood component therapy was not included in this review.
{"title":"Patient Blood Management: Anesthesiologist's Perspectives","authors":"Taehee Kim, M. Jeong","doi":"10.7599/HMR.2018.38.1.49","DOIUrl":"https://doi.org/10.7599/HMR.2018.38.1.49","url":null,"abstract":"Blood transfusion is generally considered to be the solution of anemia and blood loss during surgery. Transfusion is a very efficient and effective method to correct anemia, but there has been increasing evidence that blood transfusion does not lead to improved outcomes and that morbidity and mortality increase in a dose-dependent manner [1,2]. It has been shown that even a single unit of transfused packed red blood cells (PRBCs) can increase 30day mortality, complicated mortality, pneumonia and sepsis [3]. Therefore, it is preferable to avoid unnecessary blood transfusion or to minimize blood transfusion. In surgical patients, patient blood management focuses on anemia management, minimization of blood loss, appropriate transfusion for reducing surgical risk, and improving patient outcome after surgery. Recognition and management of pre-operative anemia represent an opportunity to optimize patient status before surgery, thereby reducing blood transfusion and potentially improving recovery from surgery and associated postoperative outcomes. A complex approach such as anesthetic strategy and operative techniques, pharmacological intervention, and cell salvage is required to reduce bleeding during surgery. In this review, we reviewed the studies about blood management in the stance of anesthesiologists. Management of coagulopathy and blood component therapy was not included in this review.","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"359 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133741432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-03-01DOI: 10.7599/HMR.2018.38.1.62
W. Lee
Obstetrics and gynecology is a subject that deals with a lot of blood. Obstetrics is also called “bloody business”. The rate of severe postpartum hemorrhage (PPH) requiring transfusion increase from 30.4 to 96.4 per 10,000 delivery hospitalizations between 1998 and 1999 to 2008 and 2009, respectively [1]. Gynecology is also closely related to blood. Large vessel injury is one of the major complications in gynecologic oncologic surgery. This is a result of massive transfusion. However, blood transfusion has potential dangerous effects which can be classified as infectious and non-infectious risks as well as immunologic causes [2]. Implications of blood transfusion occur more often in patients treated for hematologic disorder or malignancy at a rate of 1% to 6% [3,4]. Concern about viral infection such as the human immunodeficiency virus, hepatitis B and C viruses, Ebstein-Barr virus, cytomegalovirus, non A and non B hepatitis viruses are growing [5,6]. Transfusion errors contribute to non-infectious complications of Review
{"title":"Patient Blood Management: Obstetrician, Gynecologist's Perspectives","authors":"W. Lee","doi":"10.7599/HMR.2018.38.1.62","DOIUrl":"https://doi.org/10.7599/HMR.2018.38.1.62","url":null,"abstract":"Obstetrics and gynecology is a subject that deals with a lot of blood. Obstetrics is also called “bloody business”. The rate of severe postpartum hemorrhage (PPH) requiring transfusion increase from 30.4 to 96.4 per 10,000 delivery hospitalizations between 1998 and 1999 to 2008 and 2009, respectively [1]. Gynecology is also closely related to blood. Large vessel injury is one of the major complications in gynecologic oncologic surgery. This is a result of massive transfusion. However, blood transfusion has potential dangerous effects which can be classified as infectious and non-infectious risks as well as immunologic causes [2]. Implications of blood transfusion occur more often in patients treated for hematologic disorder or malignancy at a rate of 1% to 6% [3,4]. Concern about viral infection such as the human immunodeficiency virus, hepatitis B and C viruses, Ebstein-Barr virus, cytomegalovirus, non A and non B hepatitis viruses are growing [5,6]. Transfusion errors contribute to non-infectious complications of Review","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121255816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-03-01DOI: 10.7599/HMR.2018.38.1.16
H. Lee, Y. Yuh
The idea of Patient Blood Management (PBM) has emerged mainly due to problems caused by blood transfusion and perioperative anemia. This concept is based on the 5 elements suggested by Hofmann et al. [1] (2011): gaps between supply and demand for blood, high transfusion costs, risk of contaminated blood products, adverse outcomes of transfusion, and a paucity of evidence to prove transfusions efficacy. Furthermore, there is a serious issue related to perioperative anemia. The significance of managing perioperative anemia is particularly underestimated, and medical professionals use blood transfusions indiscriminately to rapidly return hemoglobin (Hb) levels to normal [2,3]. PBM is a group of multi-disciplinary protocols under the concept of 3 pillars that are applied to a patient’s clinical course (before, during and after the operation): optimizing red blood cells (RBCs) production, reducing bleeding, and harnessing the tolerance of anemia [1,4]. One of the advantages of PBM is cost-effectiveness. The Department of Health in Western Australia started comprehensive PBM; they experienced cost savings of Australian dollar (AUD) Review
{"title":"A Paradigm Shift: Perioperative Iron and Erythropoietin Therapy for Patient Blood Management","authors":"H. Lee, Y. Yuh","doi":"10.7599/HMR.2018.38.1.16","DOIUrl":"https://doi.org/10.7599/HMR.2018.38.1.16","url":null,"abstract":"The idea of Patient Blood Management (PBM) has emerged mainly due to problems caused by blood transfusion and perioperative anemia. This concept is based on the 5 elements suggested by Hofmann et al. [1] (2011): gaps between supply and demand for blood, high transfusion costs, risk of contaminated blood products, adverse outcomes of transfusion, and a paucity of evidence to prove transfusions efficacy. Furthermore, there is a serious issue related to perioperative anemia. The significance of managing perioperative anemia is particularly underestimated, and medical professionals use blood transfusions indiscriminately to rapidly return hemoglobin (Hb) levels to normal [2,3]. PBM is a group of multi-disciplinary protocols under the concept of 3 pillars that are applied to a patient’s clinical course (before, during and after the operation): optimizing red blood cells (RBCs) production, reducing bleeding, and harnessing the tolerance of anemia [1,4]. One of the advantages of PBM is cost-effectiveness. The Department of Health in Western Australia started comprehensive PBM; they experienced cost savings of Australian dollar (AUD) Review","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133785220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-11-01DOI: 10.7599/HMR.2017.37.2.61
Kyu-Hwan Jung, Hyunho Park, Woochan Hwang
Following the recent development in artificial intelligence, where deep learning has become the main methodology, the paradigm of medical image analysis is shifting from the previous clinical experience and knowledge-based feature engineering to the data-driven objective feature analysis of deep learning. Especially, as the application of various techniques developed for natural images to medical images is being accelerated, we are no longer simply adapting the natural image models to medical images but developing new methods, which encompasses the unique characteristics of the medical image domain. Furthermore, as the research on interpretability of decisions made by deep learning models and the way of incorporating clinical knowledge into the model progresses, we have started to obtain promising results that will allow clinical implementation of deep learning. Among various deep learning models, convolutional neural networks (CNN) have become methodology of choice for visual recognition problems. CNN is a type of feed-forward artificial neural network, which learns hierarchical features by iterating convolution and pooling layers until the output prediction layer is reached. While the convolution layers learn specific patterns in the input or intermediate feature map with locally-connected shared weights, pooling layers reduce the feature map by spatially aggregating activations. In special cases where the output of the model is same as the input or its denoised version, we call the model as convolutional auto-enconder (CAE). In medical image analysis, machine learning methods have been used in various fields such as detection and classification Corresponding Author: Kyu-Hwan Jung VUNO Inc., 6F, 507, Gangnamdae-ro, Seocho-gu, Seoul, Korea Tel: +82-2-515-6646 Fax: +82-2-515-6647 E-mail: kyuhwanjung@gmail.com
{"title":"Deep Learning for Medical Image Analysis: Applications to Computed Tomography and Magnetic Resonance Imaging","authors":"Kyu-Hwan Jung, Hyunho Park, Woochan Hwang","doi":"10.7599/HMR.2017.37.2.61","DOIUrl":"https://doi.org/10.7599/HMR.2017.37.2.61","url":null,"abstract":"Following the recent development in artificial intelligence, where deep learning has become the main methodology, the paradigm of medical image analysis is shifting from the previous clinical experience and knowledge-based feature engineering to the data-driven objective feature analysis of deep learning. Especially, as the application of various techniques developed for natural images to medical images is being accelerated, we are no longer simply adapting the natural image models to medical images but developing new methods, which encompasses the unique characteristics of the medical image domain. Furthermore, as the research on interpretability of decisions made by deep learning models and the way of incorporating clinical knowledge into the model progresses, we have started to obtain promising results that will allow clinical implementation of deep learning. Among various deep learning models, convolutional neural networks (CNN) have become methodology of choice for visual recognition problems. CNN is a type of feed-forward artificial neural network, which learns hierarchical features by iterating convolution and pooling layers until the output prediction layer is reached. While the convolution layers learn specific patterns in the input or intermediate feature map with locally-connected shared weights, pooling layers reduce the feature map by spatially aggregating activations. In special cases where the output of the model is same as the input or its denoised version, we call the model as convolutional auto-enconder (CAE). In medical image analysis, machine learning methods have been used in various fields such as detection and classification Corresponding Author: Kyu-Hwan Jung VUNO Inc., 6F, 507, Gangnamdae-ro, Seocho-gu, Seoul, Korea Tel: +82-2-515-6646 Fax: +82-2-515-6647 E-mail: kyuhwanjung@gmail.com","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129477990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-11-01DOI: 10.7599/HMR.2017.37.2.49
Yoonjoo Choi
{"title":"Concepts, Characteristics, and Clinical Validation of IBM Watson for Oncology","authors":"Yoonjoo Choi","doi":"10.7599/HMR.2017.37.2.49","DOIUrl":"https://doi.org/10.7599/HMR.2017.37.2.49","url":null,"abstract":"","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"613 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131473702","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-11-01DOI: 10.7599/HMR.2017.37.2.86
Yu Rang Park, S. Shin
Recently, artificial intelligence (AI) has been highlighted in various areas including healthcare [1–4]. AI can be categorized into symbolic AI such as expert systems and machine learning (ML), which includes deep learning. Technically, recently mentioned AI refers to ML or deep learning. Deep learning, which is inspired by biological neurons, is a subcategory of machine learning algorithms [5]. Machine learning (including deep learning) requires a large amount of training data to improve performance. Therefore, to implement a good healthcare AI system, we need a vast amount of healthcare data. Many people believe there is a large amount of data in hospitals based on the wide adaptation of electronic medical records (EMR). They mentioned that the adoption rate of EMR in the United States was dramatically increased to 97% after the introduction of the Health Information Technology for Economic and Clinical Health (HITECH) Act [6] and the adoption rate of EMR in Korea is more than 92%. Nearly all hospitals in Korea also use the computerized physician order entry (CPOE) system. However, the EMR adoption rate is only 58.1%, and the fully comprehensive EMR adoption rate has dropped to 11.6% [7]. This implies a lack of digitalized data for healthcare AI research in Korea. Even though there is a large amount of data, having only a large quantity of data based on big data concepts may fail to achieve an applicable healthcare AI system. We need well-curated and labeled data. For example, 54 US licensed ophthalmologists and ophthalmology senior residents have reviewed 128,175 retinal images to build a well-curated dataset [3]. Current digitalized medical records require more in-depth curation to be used for research. Moreover, to realize precision medicine with the aid of AI methods, we need many new healthcare data types including genome and wearable data. Corresponding Author: Soo-Yong Shin Department of Computer Science and Engineering, Kyung Hee University, 1732, Deogyeong-daero, Giheung-gu, Yongin-si, Gyeonggi-do 17104, Korea Tel: +82-31-201-2543 E-mail: sooyong.shin@khu.ac.kr
{"title":"Status and Direction of Healthcare Data in Korea for Artificial Intelligence","authors":"Yu Rang Park, S. Shin","doi":"10.7599/HMR.2017.37.2.86","DOIUrl":"https://doi.org/10.7599/HMR.2017.37.2.86","url":null,"abstract":"Recently, artificial intelligence (AI) has been highlighted in various areas including healthcare [1–4]. AI can be categorized into symbolic AI such as expert systems and machine learning (ML), which includes deep learning. Technically, recently mentioned AI refers to ML or deep learning. Deep learning, which is inspired by biological neurons, is a subcategory of machine learning algorithms [5]. Machine learning (including deep learning) requires a large amount of training data to improve performance. Therefore, to implement a good healthcare AI system, we need a vast amount of healthcare data. Many people believe there is a large amount of data in hospitals based on the wide adaptation of electronic medical records (EMR). They mentioned that the adoption rate of EMR in the United States was dramatically increased to 97% after the introduction of the Health Information Technology for Economic and Clinical Health (HITECH) Act [6] and the adoption rate of EMR in Korea is more than 92%. Nearly all hospitals in Korea also use the computerized physician order entry (CPOE) system. However, the EMR adoption rate is only 58.1%, and the fully comprehensive EMR adoption rate has dropped to 11.6% [7]. This implies a lack of digitalized data for healthcare AI research in Korea. Even though there is a large amount of data, having only a large quantity of data based on big data concepts may fail to achieve an applicable healthcare AI system. We need well-curated and labeled data. For example, 54 US licensed ophthalmologists and ophthalmology senior residents have reviewed 128,175 retinal images to build a well-curated dataset [3]. Current digitalized medical records require more in-depth curation to be used for research. Moreover, to realize precision medicine with the aid of AI methods, we need many new healthcare data types including genome and wearable data. Corresponding Author: Soo-Yong Shin Department of Computer Science and Engineering, Kyung Hee University, 1732, Deogyeong-daero, Giheung-gu, Yongin-si, Gyeonggi-do 17104, Korea Tel: +82-31-201-2543 E-mail: sooyong.shin@khu.ac.kr","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125991954","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-11-01DOI: 10.7599/HMR.2017.37.2.93
Tae Hyung Kim
{"title":"A Review of Deep Genomics Applying Machine Learning in Genomic Medicine","authors":"Tae Hyung Kim","doi":"10.7599/HMR.2017.37.2.93","DOIUrl":"https://doi.org/10.7599/HMR.2017.37.2.93","url":null,"abstract":"","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115526300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2017-11-01DOI: 10.7599/HMR.2017.37.2.71
Jihoon Jeong
Cancer screening in medical imaging is one of the most important areas in computerized medical software. Especially, attempts to automate the early diagnosis of cancer using computer aided detection (CAD) algorithm on chest X-ray and mammography images were the most important research topic in the field of radiology [1]. However, the results of the clinical effects of CAD are still controversial. Even there was a research about screening performance of CAD reporting that sensitivity was significantly decreased for mammograms interpreted with vs without CAD in the subset of radiologists who interpreted both with and without CAD (odds ratio, 0.53; 95% CI, 0.29-0.97) [2]. But, deep learning technology, which has recently been greatly developed, is raising expectations for the possibility of computer software related to cancer screening again. Deep learning is a kind of neural network. The neural network consists of an input layer, a hidden layer, and an output layer. Deep learning is a neural network with a large number of hidden layers. Over the past few years, deep learning has achieved tremendous performance improvements, especially in image classification [3] and speech recognition [4]. In recent Corresponding Author: Jihoon Jeong Advisor, Lunit Inc., 6th Floor, 175 Yeoksamro, Gangnam-gu, Seoul, Korea Tel: +82-10-2512-2540 E-mail: jjeong@lunit.io
{"title":"Deep Learning for Cancer Screening in Medical Imaging","authors":"Jihoon Jeong","doi":"10.7599/HMR.2017.37.2.71","DOIUrl":"https://doi.org/10.7599/HMR.2017.37.2.71","url":null,"abstract":"Cancer screening in medical imaging is one of the most important areas in computerized medical software. Especially, attempts to automate the early diagnosis of cancer using computer aided detection (CAD) algorithm on chest X-ray and mammography images were the most important research topic in the field of radiology [1]. However, the results of the clinical effects of CAD are still controversial. Even there was a research about screening performance of CAD reporting that sensitivity was significantly decreased for mammograms interpreted with vs without CAD in the subset of radiologists who interpreted both with and without CAD (odds ratio, 0.53; 95% CI, 0.29-0.97) [2]. But, deep learning technology, which has recently been greatly developed, is raising expectations for the possibility of computer software related to cancer screening again. Deep learning is a kind of neural network. The neural network consists of an input layer, a hidden layer, and an output layer. Deep learning is a neural network with a large number of hidden layers. Over the past few years, deep learning has achieved tremendous performance improvements, especially in image classification [3] and speech recognition [4]. In recent Corresponding Author: Jihoon Jeong Advisor, Lunit Inc., 6th Floor, 175 Yeoksamro, Gangnam-gu, Seoul, Korea Tel: +82-10-2512-2540 E-mail: jjeong@lunit.io","PeriodicalId":345710,"journal":{"name":"Hanyang Medical Reviews","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125076219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：