http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : Keskenler (검색결과 2 건)
- 무료
- 기관 내 무료
- 유료
-
Keskenler Mustafa Furkan,Çelik Esra,Dal Deniz 대한전기학회 2024 Journal of Electrical Engineering & Technology Vol.19 No.4
Artifi cial intelligence (AI) has signifi cantly impacted the healthcare industry, enabling the development of advanced medical devices and software that provide effi cient and precise treatments. Health 4.0, the incorporation of computing and AI technologies into healthcare, is driving the industry's digital transformation and improving the diagnosis and treatment of diseases. AI can help detect diseases such as cancer at an early stage. AI can also lower the healthcare costs by reducing the need for unnecessary biopsies and speeding up the diagnostic process. Machine learning algorithms are commonly utilized in AI-powered healthcare studies and are also used in image-based research to diagnose a variety of diseases since the integration of AI into healthcare holds great potential to improve patient care and reduce costs. In this study, we present a multi-layer machine learning (MLML) method based on the joint use of machine learning algorithms to improve the success of skin cancer diagnosis. In this respect, the MLML method with 3 layers is proposed. In the fi rst layer, decision tree, random forest, neural network, naive bayes, and support vector machine algorithms are used. After executing this layer, 5 diff erent classifi cation results are transferred to the second layer where k-nearest neighbor algorithm is utilized. In the last layer, the results are improved using the linear regression algorithm. Thanks to our method, images in the input dataset are classifi ed into three groups: cancer, not cancer, and early-stage cancer. The multi-layer architecture is utilized to make joint decisions with diff erent machine learning algorithms and remove the limitations of each algorithm so that more accurate decisions can be made. Fourteen feature extraction algorithms that were not previously used in skin cancer images are employed. Inclusion of age, gender, and region of the lesion in the decision-making process in addition to image features also contributes to obtaining better classifi cation results. The performance of the proposed method was evaluated using four metrics. The conducted experiments showed that the MLML technique achieved 88.81% accuracy, 88.89% precision, 99.17% recall, and 93.75% F1-score in classifying skin cancer images. Finally, the results were compared with other relevant studies in the literature to demonstrate the superiority of the proposed method.
-
Synthesis and properties of sol-gel derived transparent ZnO thin films: Effect of indium doping
Eyüp Fahri Keskenler,Güven Turgut 한양대학교 세라믹연구소 2016 Journal of Ceramic Processing Research Vol.17 No.12
High quality and transparent indium doped ZnO (IZO) thin films were deposited on glass substrates by sol-gel spin coatingmethod. Zinc acetate and indium (III) chloride were used as precursor solution materials. Structural, morphological, andoptical properties of the films were investigated as a function of indium doping ranging from 0.5% to 2.0 at % by X-raydiffraction, scanning electron microscopy, transmission and energy dispersive X-ray techniques. The films had polycrystallinenature and exhibited a hexagonal wurtzite structure with preferred c-axis orientation. The film surfaces exhibited uniformparticle-like and granular morphologies. The optical transmittance spectra of the undoped ZnO and IZO films were taken inthe wavelength ranging from 350-1000 nm. The transmittance IZO films compared to undoped ZnO has increased withincreasing indium content. The chemical composition of the films indicated the presence of indium element in the ZnO films. These results make IZO thin films an attractive candidate for transparent material applications such as solar cells.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
