• Journal of Information Processing Systems
• /
• 제18권3호
• /
• pp.302-310
• /
• 2022

Efficient use of limited blood products is becoming very important in terms of socioeconomic status and patient recovery. To predict the appropriateness of patient-specific transfusions for the intensive care unit (ICU) patients who require real-time monitoring, we evaluated a model to predict the possibility of transfusion dynamically by using the Medical Information Mart for Intensive Care III (MIMIC-III), an ICU admission record at Harvard Medical School. In this study, we developed an explainable machine learning to predict the possibility of red blood cell transfusion for major medical diseases in the ICU. Target disease groups that received packed red blood cell transfusions at high frequency were selected and 16,222 patients were finally extracted. The prediction model achieved an area under the ROC curve of 0.9070 and an F1-score of 0.8166 (LightGBM). To explain the performance of the machine learning model, feature importance analysis and a partial dependence plot were used. The results of our study can be used as basic data for recommendations related to the adequacy of blood transfusions and are expected to ultimately contribute to the recovery of patients and prevention of excessive consumption of blood products.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권3호
• /
• pp.974-992
• /
• 2021

Recently, the healthcare field has undergone rapid changes owing to the accumulation of health big data and the development of machine learning. Data mining research in the field of healthcare has different characteristics from those of other data analyses, such as the structural complexity of the medical data, requirement for medical expertise, and security of personal medical information. Various methods have been implemented to address these issues, including the machine learning model and cloud platform. However, the machine learning model presents the problem of opaque result interpretation, and the cloud platform requires more in-depth research on security and efficiency. To address these issues, this paper presents a recent technology for Internet-of-Things-based (IoT-based) health big data processing. We present a cloud-based IoT health platform and health big data processing technology that reduces the medical data management costs and enhances safety. We also present a data mining technology for health-risk prediction, which is the core of healthcare. Finally, we propose a study using explainable artificial intelligence that enhances the reliability and transparency of the decision-making system, which is called the black box model owing to its lack of transparency.

• International Journal of Internet, Broadcasting and Communication
• /
• 제16권3호
• /
• pp.146-157
• /
• 2024

Despite numerous air mattresses marketed to prevent Pressure Ulcers (PU), none have fully succeeded due to residual pressure surpassing critical levels. We introduces an innovative medical bed system aiming at complete PU prevention. This system employs a unique 4-bar link mechanism, moving keys up and down to manage body pressure. Each of the 17 keys integrates a sensor controller, reading pressure from 10 sensors. By regulating motor input, we maintain body pressure below critical levels. Keys are equipped with a servo drive and sensor controller, linked to the main controller via two CAN series. Using fuzzy or PI/IP controllers, we adjust keys to minimize total error, dispersing body pressure and ensuring comfort. In case of controller failure, keys alternate swiftly, preventing ulcer development. Through experimental tests under varied conditions, the fuzzy controller with tailored membership functions demonstrated swift performance. PI control showed rapid convergence, while IP control exhibited slower convergence and oscillations near zero error. Our specialized medical robot bed, incorporating 4-bar links and pressure sensors, underwent testing with three controllers-fuzzy, PI, and IP-showcasing their effectiveness in keeping body pressure below critical ulcer levels. Experimental results validate the proposed approach's efficacy, indicating potential for complete PU prevention.

• Asian Pacific Journal of Cancer Prevention
• /
• 제14권9호
• /
• pp.5513-5518
• /
• 2013

IFN-${\gamma}$ plays an indirect anti-cancer role through the immune system but may have direct negative effects on cancer cells. It regulates the viability of gastric cancer cells, so we examined whether it affects their proliferation and how that might be brought about. We exposed AGS, HGC-27 and GES-1 gastric cancer cell lines to IFN-${\gamma}$ and found significantly reduced colony formation ability. Flow cytometry revealed no effect of IFN-${\gamma}$ on apoptosis of cell lines and no effect on cell aging as assessed by ${\beta}$-gal staining. Microarray assay revealed that IFN-${\gamma}$ changed the mRNA expression of genes related to the cell cycle and cell proliferation and migration, as well as chemokines and chemokine receptors, and immunity-related genes. Finally, flow cytometry revealed that IFN-${\gamma}$ arrested the cells in the G1/S phase. IFN-${\gamma}$ may slow proliferation of some gastric cancer cells by affecting the cell cycle to play a negative role in the development of gastric cancer.

• Perinatology
• /
• 제29권4호
• /
• pp.189-194
• /
• 2018

Congenital neuroblastoma is a rare disease. Placental metastasis is extremely rare and poor prognosis has been reported in neonates. Mirror syndrome could occur in mother with placental metastasis with possibilities of hypertension and edema. We report a case of detection of left suprarenal mass in fetus at $31^{+5}weeks^{\prime}$ gestation. Mother presented with palpitation, edema, headache, and hypertension. Maternal 24 hours urine vanillylmandelic acid (VMA) and normetanephrine (NME) level at 34 weeks' gestation were elevated. Consequently, emergent cesarean section was done. Based on abdominal ultrasonography and whole body magnetic resonance imaging, left adrenal tumor with liver metastasis was suspected. Neuroblastoma was confirmed by liver and placenta biopsy. Chemotherapy was started with Pediatric Oncology Group 9243 at day 7 and changed into Children's Oncology Group 3961 due to cholestasis and poor response during 2nd cycle. Plasma exchange was done for aggravated direct hyperbilirubinemia. The baby expired at 73 days due to multi-organ failure. Maternal symptoms were completely resolved in 2 weeks after delivery along with normalization of the elevated level of VMA and NME. We report a first case of mirror syndrome in Korean mother and fetus resulting from metastatic congenital neuroblastoma to placenta.

• 문화기술의 융합
• /
• 제10권2호
• /
• pp.453-458
• /
• 2024

광학 문자 인식(Optical Character Recognition, OCR)은 이미지 내의 문자를 인식하여 디지털 포맷(Digital Format)의 텍스트로 변환하는 기술이다. 딥러닝(Deep Learning) 기반의 OCR이 높은 인식률을 보여줌에 따라 대량의 기록 자료를 보유한 많은 산업 분야에서 OCR을 활용하고 있다. 특히, 의료 산업 분야는 의료 서비스 향상을 위해 딥러닝 기반의 OCR을 적극 도입하였다. 본 논문에서는 딥러닝 기반 OCR 엔진(Engine) 및 의료 데이터에 특화된 OCR의 동향을 살펴보고, 의료 OCR의 발전 방향에 대해 제시한다. 현재의 의료 OCR은 검출한 문자 데이터를 자연어 처리(Natural Language Processing, NLP)하여 인식률을 개선하였다. 그러나, 정형화되지 않은 손글씨(Handwriting)나 변형된 문자에서는 여전히 인식 정확도에 한계를 보였다. 의료 데이터의 데이터베이스(Database)화, 이미지 전처리(Pre-processing), 특화된 자연어 처리를 통해 더욱 고도화된 의료 OCR을 발전시키는 것이 필요하다.

• Biomolecules & Therapeutics
• /
• 제32권5호
• /
• pp.582-600
• /
• 2024

Tyrosine kinase inhibitors (TKIs) have revolutionized the treatment landscape for chronic myeloid leukemia (CML). However, TKI resistance poses a significant challenge, leading to treatment failure and disease progression. Resistance mechanisms include both BCR::ABL1-dependent and BCR::ABL1-independent pathways. The mechanisms underlying BCR::ABL1 independence remain incompletely understood, with CML cells potentially activating alternative signaling pathways, including the AKT/mTOR and JAK2/STAT5 pathways, to compensate for the loss of BCR::ABL1 kinase activity. This study explored tumoral VISTA (encoded by VSIR) as a contributing factor to TKI resistance in CML patients and identified elevated tumoral VISTA levels as a marker of resistance and poor survival. Through in vitro and in vivo analyses, we demonstrated that VSIR knockdown and the application of NSC-622608, a novel VISTA inhibitor, significantly impeded CML cell proliferation and induced apoptosis by attenuating the AKT/mTOR and JAK2/STAT5 pathways, which are crucial for CML cell survival independent of BCR::ABL1 kinase activity. Moreover, VSIR overexpression promoted TKI resistance in CML cells. Importantly, the synergistic effect of NSC-622608 with TKIs offers a potent therapeutic avenue against both imatinib-sensitive and imatinib-resistant CML cells, including those harboring the challenging T315I mutation. Our findings highlight the role of tumoral VISTA in mediating TKI resistance in CML, suggesting that inhibition of VISTA, particularly in combination with TKIs, is an innovative approach to enhancing treatment outcomes in CML patients, irrespective of BCR::ABL1 mutation status. This study not only identified a new pathway contributing to TKI resistance but also revealed the possibility of targeting tumoral VISTA as a means of overcoming this significant clinical challenge.

• 한국융합학회논문지
• /
• 제12권4호
• /
• pp.9-21
• /
• 2021

본 연구는 IP5국가(KR, EP, JP, US, CN)의 바이오헬스 분야 특허데이터를 기반으로 기술의 융합과 트렌드를 파악하여 해당 산업 분야의 발전 방향을 제시하는 것을 목적으로 한다. 기술융합 현황 파악을 위해 특허 동시분류분석 기반의 네트워크분석과 TF-IDF 기반의 텍스트마이닝을 주요 방법론으로 활용하였고, 분석 결과 바이오헬스 산업의 기술융합 클러스터는 크게 (A)치료용 의료기기, (B)의료데이터프로세싱, (C)생체계측용 의료기기의 세 가지 형태로 도출되었다. 또한 기술융합 결과를 토대로 한 트렌드 분석의 결과에서 우리나라는 (B)의료데이터프로세싱 분야에서 시장선도국으로 도출됨에 따라 향후 상업적 가치가 높은 특허로 시장 우위를 선점할 수 있는 가능성이 높다고 분석되었다. 특히 해당 분야는 2019년 1월 국회에서 통과된 '데이터3법'이라는 정책적 변환과 더불어, 국내 바이오헬스 기업들의 의료데이터 활용 가능성이 확대됨에 따라 해당 기술에 대한 기술융합 활성화 정책 수립과 R&D 지원 전략이 필요할 것으로 전망된다.

• 한국산업보건학회지
• /
• 제34권3호
• /
• pp.202-213
• /
• 2024

Introduction: Acknowledging the global issue of diseases potentially caused by overwork, this study aims to develop an AI model to help workers understand the connection between cerebrocardiovascular diseases and their work environment. Materials and methods: The model was trained using medical and legal expertise along with data from the 2021 occupational disease adjudication certificate by the Industrial Accident Compensation Insurance and Prevention Service. The Polyglot-ko-5.8B model, which is effective for processing Korean, was utilized. Model performance was evaluated through accuracy, precision, sensitivity, and F1-score metrics. Results: The model trained on a comprehensive dataset, including expert knowledge and actual case data, outperformed the others with respective accuracy, precision, sensitivity, and F1-scores of 0.91, 0.89, 0.84, and 0.87. However, it still had limitations in responding to certain scenarios. Discussion: The comprehensive model proved most effective in diagnosing work-related cerebrocardiovascular diseases, highlighting the significance of integrating actual case data in AI model development. Despite its efficacy, the model showed limitations in handling diverse cases and offering health management solutions. Conclusion: The study succeeded in creating an AI model to discern the link between work factors and cerebrocardiovascular diseases, showcasing the highest efficacy with the comprehensively trained model. Future enhancements towards a template-based approach and the development of a user-friendly chatbot webUI for workers are recommended to address the model's current limitations.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 추계학술발표대회
• /
• pp.520-522
• /
• 2022

의료 인공지능은 특정 진단에서 높은 정확도를 보이지만 모델의 신뢰성 문제로 인해 활발하게 쓰이지 못하고 있다. 이에 따라 인공지능 모델의 진단에 대한 원인 설명의 필요성이 대두되었고 설명가능한 의료 인공지능에 관한 연구가 활발히 진행되고 있다. 하지만 MRI 등 의료 영상 인공지능 분야에서 주로 진행되고 있으며, 이미지 형태가 아닌 전자의무기록 데이터 (Electronic Health Record, EHR) 를 기반으로 한 모델의 설명가능성 연구는 EHR 데이터 자체의 복잡성 때문에 활발하게 진행 되지 않고 있다. 본 논문에서는 전자의무기록 데이터인 MIMIC-III (Medical Information Mart for Intensive Care) 를 전처리 및 그래프로 표현하고, GCT (Graph Convolutional Transformer) 모델을 학습시켰다. 학습 후, 어텐션 흐름 그래프를 시각화해서 모델의 예측에 대한 직관적인 설명을 제공한다.