CD Skripsi

Implementasi Deep Learning Image Recognition Untuk Deteksi Penyakit Kulit Pada Manusia Menggunakan Convolutional Neural Network (Cnn)

---

Skin disease is the worldwide most common health problem, especially in Indonesia. Globally, there are over 200 million people affected by one form of skin disease, with a majority occurring in developing countries with hot tropical climates. In Indonesia, skin diseases placed on rank third among the top ten diseases in the outpatient category. Diagnosis of skin diseases can often be harder for laypersons and sometimes requires a long wait for accurate results. This research proposes the use of Convolutional Neural Network (CNN) algorithms for skin disease detection, which can aid in faster and more accurate diagnosis, so can get timely treatment and management. CNN is implemented using a diverse dataset of skin disease images, preprocessing steps to enhance image quality, training and testing processes, and periodic model evaluation. Evaluation is performed by measuring accuracy, precision, recall, and F1-score, as well as observing graphs of training results and the confusion matrix. The results obtained in this study show that the CNN model performs optimally using 1000 skin disease image datasets with 4 disease categories, a pre-trained VGG19 model, 2 additional layers using transfer learning approach, achieving an accuracy of 98%, precision of 98%, recall of 98%, and an F1-score of 98%. In conclusion, the research successfully maximized the CNN model into its best performance.

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2003113241

Penerbit

Pekanbaru : Universitas Riau FMIPA Sistem Informasi., 2024

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2003113241

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

SISTEM INFORMASI

Info Detail Spesifik

-

Pernyataan Tanggungjawab

ERA

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar