• 디지털융복합연구
• /
• 제16권1호
• /
• pp.231-242
• /
• 2018

병원 재원일수의 효율적 관리는 병원의 수익과 환자의 진료비 절감을 위해 매우 중요한 요소이다. 이러한 재원일수의 효율적 관리를 위해서는 병원들이 재원일수에 대해서 벤치마킹을 할 수 있도록 지원이 필요하고 재원일수 절감의 구체적인 방향을 제시해 줄 수 있는 재원일수 예측모형의 개발이 필요하다. 본 연구에서는 2013년과 2014년도 퇴원손상환자자료 중 급성뇌졸중 환자를 추출하여 분석용 자료를 만들고 인공지능을 이용하여 급성뇌졸중 환자의 재원일수 예측모형을 개발하였다. 분석용 자료는 훈련용 60%, 평가용 40%로 분류하였다. 모형개발은 전통적 통계기법인 다중회귀분석기법과 인공지능기법인 대화식 의사결정나무기법, 신경망 기법, 그리고 이들을 모두 통합한 앙상블기법을 이용하였다. 모형평가는 Root ASE(Absolute error) 지표를 이용하였는데, 다중회귀분석은 23.7, 대화식결정나무 23.7, 신경망 분석은 22.7, 앙상블은 22.7로 나타났고 이를 통하여 재원일수 예측모형 개발에 인공지능기법의 유용성이 입증되었다. 앞으로 재원일수 예측모형개발에 인공지능 기법을 보다 효율적으로 활용할 수 있는 방안에 대해서 계속적인 연구가 이루어 질 필요가 있다.

• Tuberculosis and Respiratory Diseases
• /
• 제73권5호
• /
• pp.282-287
• /
• 2012

Broncholiths are defined as calcified materials that occur in a tracheobronchial tree or in a cavity communicating with that. Broncholith has variable clinical features. The therapeutic options to remove broncholiths are so variable that clinicians need to select the most safe and effective methods by mass size, mobility, and location. As yet, there is no consistent guideline removing a broncholith. We report 2 successful cases of removing a fixed broncholith by flexible bronchoscopy guided cryoadhesion. With repeated technique of thawing and freezing with cryoprobe, we could extract the fixed broncholith safely. This method is promising as a way to remove broncholith in the future.

• 대한두경부종양학회지
• /
• 제33권2호
• /
• pp.75-79
• /
• 2017

Bronchogenic cysts are congenital malformations of the bronchial tree, a type of bronchopulmonary foregut malformation. The presentation of the bronchogenic cyst is variable, making pre-operative diagnosis difficult. They aremostly asymptomatic orarefound incidentally when the chest is imaged. They can present as lower neck massesor mediastinal masses that may enlarge. They cause mass effect due to local compression and may result in tracheo-bronchial obstruction leading to air trapping and respiratory distress. The treatment is somewhat controversial, and in general,these lesions are treated using the transcervical or transbronchial approach. When these cysts arelocalized in the upper mediastinum, it may be possible to removethemusing the transcervical approach. In our three cases, the patientscomplained of mild dysphagia, foreign body sensation, and dyspnea. We report three cases of a large bronchogenic cyst in the lower neck and the upper mediastinum treated using the transcervical approach.

• 의료법학
• /
• 제9권1호
• /
• pp.165-196
• /
• 2008

Recently the Supreme Court held that "in cases in which a patient suffering from a mental disorder attempts to commit suicide, fails, and then succeeds in a subsequent attempt, the following circumstances must be present in order to acknowledge reasonable causation between the negligence of the hospital with regards to taking care of the patient and the death of the patient; there must have existed negligence on the part of the hospital with regards to their failure to stop the 1stsuicide attempt, injurious aftereffects must have been caused to the patient by the1stsuicide attempt, and said aftereffects must have been the main cause for the 2nd successful suicide attemtp." This, in effect, lessens the requirements of past holdings of the Supreme Court which held that "to acknowledge reasonable causation between the negligence of the hospital and the patient that commits suicide, the patient must have experienced such severe physical and mental suffering from the previous attempt so that they could not help but choose to commit suicide". The fact that the Supreme Court did not clearly state such changes in their view on this matter should be corrected. Also, the fact that the court only held the hospital liable for damages of less than 50 million won, only calculating damages up to the point when the deceased passed, is inadequate compared to other cases and should be corrected.

• Archives of Plastic Surgery
• /
• 제48권6호
• /
• pp.646-650
• /
• 2021

Head and neck reconstruction poses unique challenges in rehabilitating surgical defects in terms of integrity, function, and form. The radial forearm free flap (RFFF) has been widely used for defect coverage, especially in the head and neck area, but its versatility allows it to be used for soft-tissue reconstruction in various parts of the body. The vascular features of the flap are quite constant and reliable. Nevertheless, abnormalities of the forearm vascular tree have been described over the decades. We report a case of intraoral reconstruction after verrucous carcinoma recurrence in a 74-year-old woman with an unusual forearm flap, which we called the median forearm free flap, based on a median branch of the radial artery that was preoperatively detected using handheld Doppler ultrasonography. The distally located skin paddle was predominantly supplied by the aberrant median vessel with its perforators. The flap was thus safely harvested with this atypical pedicle. Successful reconstruction of the intraoral defect was achieved, with an uneventful postoperative course.

• 대한영상의학회지
• /
• 제82권1호
• /
• pp.207-211
• /
• 2021

기관지 낭종은 기관이나 기관지를 따라 종격동에서 가장 흔하게 나타나는 드문 선천성 기형이며, 특히 뒤쪽 상부 종격동에 생긴다. 기관지 낭종은 다른 위치에서도 발생할 수 있으며 폐, 복강, 그리고 후복강에서 위치가 보고되었다. 저자들은 전천추부 공간에 생긴 후복강 기관지 낭종의 증례를 보고한다. 환자는 수술 후 병리 조직학적 검사에서 최종적으로 진단되었다.

• 대한간호학회지
• /
• 제40권3호
• /
• pp.359-366
• /
• 2010

Purpose: The purpose of this study was to determine the significant factors for risk estimate of aspiration and to evaluate the efficiency of the dysphagia assessment tool. Methods: A consecutive series of 210 stroke patients with aspiration symptoms such as cough and dysphagia who had soft or regular diet without tube feeding were examined. The dysphagia assessment tool for aspiration was compared with videofluoroscopy using Classification and Regression Tree (CART) analysis. Results: In CART analysis, of 34 factors, the significant factors for estimating risk of aspiration were cough during swallowing, oral stasis, facial symmetry, salivary drooling, and cough after swallowing. The risk estimate error of the revised dysphagia assessment tool was 25.2%, equal to that of videofluoroscopy. Conclusion: The results indicate that the dysphagia assessment tool developed and examined in this study was potentially useful in the clinical field and the primary risk estimating factor was cough during swallowing. Oral stasis, facial symmetry, salivary drooling, cough after swallowing were other significant factors, and based on these results, the dysphagia assessment tool for aspiration was revised and complemented.

• Tuberculosis and Respiratory Diseases
• /
• 제74권5호
• /
• pp.231-234
• /
• 2013

Aspergillus causes a variety of clinical syndromes in the lung including tracheobronchial aspergillosis, invasive aspergillosis, chronic necrotizing pulmonary aspergillosis, allergic bronchopulmonary aspergillosis, and aspergilloma. Aspergilloma usually results from ingrowths of colonized Aspergillus in damaged bronchial tree, pulmonary cyst or cavities of patients with underlying lung diseases. There are a few reports on endobronchial aspergilloma without underlying pulmonary lesion. We have experienced a case of endobronchial aspergilloma associated with foreign body developed in an immunocompetent patient without underlying lung diseases. A 59-year-old man is being hospitalized with recurring hemoptysis for 5 months. X-ray and computed tomography scans of chest showed a nodular opacity in superior segment of left lower lobe. Fiberoptic bronchoscopy revealed an irregular, mass-like, brownish material which totally obstructed the sub-segmental bronchus and a foreign body in superior segmental bronchus of the lower left lobe. Histopathologic examinations of biopsy specimen revealed fungal hyphae, characteristic of Aspergillus species.

• 한국병원경영학회지
• /
• 제13권4호
• /
• pp.27-45
• /
• 2008

This research focused on a total of 378 patients with external injuries under the age of 14 who visited the emergency center at a university hospital in Korea, during the months of January, April, August and November between January and December of 2007. In addition, the survey was conducted only on even number days and ranged across the whole year to reflect the impact of seasonal characteristics on the collected data. The research focused on determining the characteristics and inflicting cause of these patients with external injuries, analyzed the total time spent in the emergency room and obtained the following results. 1. When classifying the patients into different genders, the proportion of males (67.5%) was higher than that of females (32.5%). According to the different age groups, the highest ratio, at 61.1 %, was patients under the age of six. 2. Looking at the total number of minutes spent in the emergency room, the longest amount of time occurred during April with 162.7 minutes, followed by 121.9 minutes in January and 92.4 minutes in November. August had the shortest period of time spent in the emergency room, a significant statistical difference from the other periods of the year (p<0.001). 3. Regarding the amount of time required for each examination, patients required to provide a urine test spent an average of 204.7 minutes while those who did not spent 113.5 minutes on average. This is a 5% statistical difference among the two groups (p>0.05). 4. Looking at the five most commonly diagnosed problems in the emergency room, the total number of people with these top five commonly diagnosed illnesses comprised 55.6%, or 210 patients out of 378. 5. Utilizing the Decision Tree Model to estimate the total number of minutes required per visit, the first classifications were made using a chemical examination factor. People subject to chemical classification spent an average of 177.7 minutes, which was longer than the overall average of 115.2 minutes, and those exempt from chemical examination spent an average of 103.8 minutes, which was shorter than the average Conclusion; Effort to curtail the total time spent in emergency rooms is vital in guaranteeing efficient management of hospitals and providing medical services. The delay experienced by many comprehensive professional medical centers must be resolved through the establishment of effective delivery of medical services, increased supply of patient rooms and other policy oriented implementations. However, for now, this problem must be resolved by increasing the level of patient satisfaction and guaranteeing effective operation of patient rooms, which will significantly contribute to the general management and success of hospitals and institutions.

• Korean Journal of Radiology
• /
• 제22권3호
• /
• pp.344-353
• /
• 2021

Objective: The mitotic count of gastrointestinal stromal tumors (GIST) is closely associated with the risk of planting and metastasis. The purpose of this study was to develop a predictive model for the mitotic index of local primary GIST, based on deep learning algorithm. Materials and Methods: Abdominal contrast-enhanced CT images of 148 pathologically confirmed GIST cases were retrospectively collected for the development of a deep learning classification algorithm. The areas of GIST masses on the CT images were retrospectively labelled by an experienced radiologist. The postoperative pathological mitotic count was considered as the gold standard (high mitotic count, > 5/50 high-power fields [HPFs]; low mitotic count, ≤ 5/50 HPFs). A binary classification model was trained on the basis of the VGG16 convolutional neural network, using the CT images with the training set (n = 108), validation set (n = 20), and the test set (n = 20). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated at both, the image level and the patient level. The receiver operating characteristic curves were generated on the basis of the model prediction results and the area under curves (AUCs) were calculated. The risk categories of the tumors were predicted according to the Armed Forces Institute of Pathology criteria. Results: At the image level, the classification prediction results of the mitotic counts in the test cohort were as follows: sensitivity 85.7% (95% confidence interval [CI]: 0.834-0.877), specificity 67.5% (95% CI: 0.636-0.712), PPV 82.1% (95% CI: 0.797-0.843), NPV 73.0% (95% CI: 0.691-0.766), and AUC 0.771 (95% CI: 0.750-0.791). At the patient level, the classification prediction results in the test cohort were as follows: sensitivity 90.0% (95% CI: 0.541-0.995), specificity 70.0% (95% CI: 0.354-0.919), PPV 75.0% (95% CI: 0.428-0.933), NPV 87.5% (95% CI: 0.467-0.993), and AUC 0.800 (95% CI: 0.563-0.943). Conclusion: We developed and preliminarily verified the GIST mitotic count binary prediction model, based on the VGG convolutional neural network. The model displayed a good predictive performance.