Claim Missing Document
Check
Articles

Found 21 Documents
Search

Klasifikasi Jenis Penyakit Tumor Otak Menggunakan Convolutional Neural Network Arsitektur Inception-V3 Hendra Ady Juliartadi; Andi Kurniawan Nugroho; Sri Heranurweni; Titik Nurhayati
Electrician : Jurnal Rekayasa dan Teknologi Elektro Vol. 19 No. 3 (2025)
Publisher : Department of Electrical Engineering, Faculty of Engineering, Universitas Lampung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23960/elc.v19n3.2868

Abstract

Tumor otak merupakan salah satu penyakit yang paling mengancam nyawa dan memerlukan diagnosis yang akurat. Salah satu tantangan utama dalam mendiagnosis tumor otak terletak pada kesamaan visual antara jenis tumor pada gambar MRI, yang seringkali menyulitkan klasifikasi manual. Untuk mengatasi hal ini, sistem klasifikasi berbasis deep learning dikembangkan menggunakan metode Convolutional Neural Network (CNN) dengan arsitektur Inception-V3. Data yang digunakan diperoleh dari Kaggle dan terdiri dari empat kelas glioma, meningioma, pituitari, dan notumor. Penelitian ini menerapkan teknik pre-procesing seperti Contrast Limited Adaptive Histogram Equalization (CLAHE) dan augmentasi data untuk meningkatkan kinerja model. Berbagai kombinasi hyperparameters diuji, dan kinerja terbaik dicapai dengan batch size 32 dan learning rate 0,0001. Model mencapai accuracy 97,92%, precision 98%, recall 98%, dan F1-score 98%. Selain itu, arsitektur Inception-V3 dibandingkan dengan arsitektur lain yaitu MobileNetV2 dan ResNet50, yang mencapai accuracy 83,67% dan 71,79%. Hasil menunjukkan bahwa arsitektur Inception-V3 unggul dalam mengklasifikasikan jenis tumor otak dari gambar MRI. Penelitian ini diharapkan dapat berkontribusi dalam meningkatkan akurasi dan kecepatan diagnosis tumor otak bagi medis.
Implementasi Teknologi Penggilingan Padi Portabel Berbasis IoT untuk Meningkatkan Efisiensi Pasca Panen di Desa Tamangede Mudjiastuti Handajani; Andi Kurniawan Nugroho; Muhammad Ikhsan Setiawan
Archive: Jurnal Pengabdian Kepada Masyarakat Vol. 5 No. 2 (2026): Juni 2026
Publisher : Asosiasi Pengelola Publikasi Ilmiah Perguruan Tinggi PGRI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55506/arch.v5i2.265

Abstract

Latar Belakang: Petani padi di Desa Tamangede menghadapi permasalahan efisiensi rendah dalam pengolahan pasca panen akibat ketergantungan pada penggilingan konvensional yang menimbulkan biaya tinggi dan losses mencapai 15-20%. Tujuan: mengimplementasikan teknologi Triple Function Electrical Portable Mini Rice Milling (TFEPMRM) berbasis IoT untuk meningkatkan efisiensi pengolahan gabah dan pendapatan petani. Metode: Pengabdian dilakukan melalui pendekatan partisipatif dengan transfer teknologi, pendampingan pembuatan mesin, pelatihan operasional, dan monitoring berbasis IoT pada kelompok tani selama periode Januari hingga Oktober 2025. Hasil: Implementasi menghasilkan peningkatan efisiensi dengan kapasitas 80 kg/6 menit, efisiensi dehusking 95%, milling recovery 83.6-87.4%, pengurangan biaya operasional 84.2%, dan potensi peningkatan pendapatan petani mencapai Rp 5.427.500 per tahun per hektar. Kesimpulan: Teknologi TFEPMRM efektif meningkatkan efisiensi pengolahan padi dan berpotensi meningkatkan pendapatan petani melalui pengurangan losses dan biaya operasional.
Pengaruh Desain Modular Terhadap Stabilitas Suhu Dan Reduksi Kebisingan Pada Incu Analyzer Fajri Prasojo; Andi Kurniawan Nugroho; Sri Heranurweni
Jurnal Transformatika Vol. 23 No. 2 (2026): January 2026
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v23i2.12330

Abstract

Incu analyzer merupakan alat yang digunakan untuk mengukur suhu, kebisingan aliran udara dan kelembaban. Cara kerja alat pada saat pengkalibrasian diletakan di atas matras. Sensor suhu yangdigunakan adalah sensor LM35 yang menghasilkan nilai tegangan linier dengan suhu. Tengangan keluaranya sebanding dengan suhu derajat celcius. Pada sensor tingkat kebisingan menggunakan sound sensor microphone module yaitu modul elektronik yang berfungsi untuk mendeteksi suara dan mengubah menjadi sinyal listrik. Sound Mic Microphone Sensor Detection Module for Arduino ini mempunya range pendeteksi suara ±30 cm hingga 1 meter. Pada saat pengujian setting 35°C di dapatkan hasil presentase error sebesar 7%, setting 36°C didapatkan hasil presentase error sebesar 3% dan setting 37°C hasil presentase error sebesar 3%. Data pengujian sensor tingkat kebisingan dilakukan sebanyak 9 kali pada sela waktu 5 menit. Dan didapatkan rata rata nilai error sebesar 1,2(dB).
RANCANG BANGUN AUTOMATIC RAIN RECORDING (ARR ) BERBASIS DEVELOPMENT BOARD MIKROAVR128 Andi Kurniawan Nugroho; Edy Susilo; Diah Setyati Budiningrum
Elektrika Vol. 9 No. 1 (2017): April 2017
Publisher : Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (635.43 KB) | DOI: 10.26623/elektrika.v9i1.1107

Abstract

 The right use of rain can provide many benefits. For this reason, mapping of rainfall areas is needed. This is intended to determine the level of disaster preparedness for each region. In mapping, a tool is needed to calculate the rainfall that falls. In the market there are only manual measuring devices, while rainfall gauges are not available which can measure automatically and directly but additional equipment is needed separately. In addition, the measurement results cannot be seen by the public. Therefore we need a tool that works automatically that can display measurement results directly and can be seen directly by the public. Thus the community can see and measure rainfall that falls. In this plan Automatic Rain Recording is used by using a proximity sensor placed on top of a measuring cup and the sensor is connected to the AVR128 Development Board as the control center to convert sensor data into ADC (Analog to Digital Converter) data. This tool is connected with a microSD with a capacity of 8Gbyte to store rainfall level data. AVR128 will store every 2 minutes with a total of 275,368 data and 382 days of storage capability.Keywords : rainfall, ARR, ADC, AVR128, mikroSD  
RANCANG BANGUN ROBOT PEMADAM API BERODA DENGAN NAVIGASI SENSOR KOMPAS BERBASIS ATMEGA 128 Yusuf Nurul hilal; Sri Heranurweni; Andi Kurniawan N
Elektrika Vol. 9 No. 1 (2017): April 2017
Publisher : Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (527.142 KB) | DOI: 10.26623/elektrika.v9i1.1108

Abstract

CMPS03 an electronic compass sensor navigation sensor that helps in determining the direction toward the robot and in tracing the wall or room that works according to the principle of the compass in general. CMPS03 electronic compass sensor has two sensors that will detect ferrous magnetic Earth's magnetic field and also has a range of angles 360Ëš, the resulting data is then sent to the microcontroller Keywords: Compass Sensor, Robot, and Microcontroller.
Analisa Citra Medis Pada Pasien Stroke dengan Metoda Peregangan Kontras Berbasis ImageJ Budiani Destyningtias; Andi Kurniawan N; Sri Heranurweni
Elektrika Vol. 10 No. 1 (2018): April 2018
Publisher : Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (999.936 KB) | DOI: 10.26623/elektrika.v10i1.1130

Abstract

Penelitian ini bertujuan mengembangankan teknologi pengolahan citra medis terutama citra medis CT Scan penderita stroke. Dokter dalam menentukan tingkat keparahan pasien stroke biasanya menggunakan citra medis CT scan dan mengalami kesulitan dalam menginterpretasikan luasan cakupan perdarahan.Solusi yang digunakan dengan peregangan kontras yang akan membedakan jaringan sel, tulang tengkorak dan jenis perdarahan. Penelitian ini menggunakan peregangan kontras dari hasil citra CT Scan yang dihasilkan dengan terlebih dahulu mengubah Citra DICOM menjadi citra JPEG dengan menggunakan bantuan program ImageJ. Hasil penelitian menunjukkan bahwa metode histogram ekualisasi dan analisis tekstur statistik dapat digunakan untuk membedakan yang normal MRI dan abnormal MRI yang terdeteksi stroke. Kata Kata Kunci : Stroke, MRI , Dicom, JPEG, ImageJ, Peregangan Kontras
KLASIFIKASI POLA IMAGE PADA PASIEN TUMOR OTAK BERBASIS JARINGAN SYARAF TIRUAN ( STUDI KASUS PENANGANAN KURATIF PASIEN TUMOR OTAK ) Sri Heranurweni; Budiani Destyningtias; Andi Kurniawan Nugroho
Elektrika Vol. 10 No. 2 (2018): Oktober 2018
Publisher : Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (733.053 KB) | DOI: 10.26623/elektrika.v10i2.1169

Abstract

 Nowadays medical science has developed rapidly, diagnostic and treatment techniques have provided life expectancy for patients. One way of examining brain tumor sufferers is radiological examination that needs to be done, including MRI with contrast. MRI brain images are useful for seeing tumors in the initial steps of diagnosis and are very good for classification, erosions / destruction lesions of the skull. Smoothing image processing, segmentation with otsu method and feature extraction are carried out to facilitate the training and testing process. This study, will apply texture analysis with the parameters contrast, correlation, energy, homogenity to distinguish the texture of brain tumor images and normal so as to produce a standard gold value based on existing texture characteristics. Training and testing of texture features using backpropagation method of artificial neural networks with variations in learning rate values so that it is expected to obtain a classification of the image conditions of patients with brain tumors. The data used are 29 brain images that produce classification accuracy of 96.55%.Keywords :   MRI images, brain tumors, textur, backprogation  
PENGATURAN SUHU UNTUK MENGATASI BAYI TERBAKAR BERBASIS ARDUINO DAN LABVIEW PADA INFANT INCUBATOR Dicky Rivaldo Ramdani; Andi Kurniawan Nugroho; Budiani Destyningtias
Elektrika Vol. 11 No. 1 (2019): April 2019
Publisher : Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (688.546 KB) | DOI: 10.26623/elektrika.v11i1.1540

Abstract

An incubator is a place designed to maintain a certain temperature condition. Incubators are often found in hospitals and farms. In hospitals, the Incubator functions to warm newborns, or premature babies. On farms, this incubator is usually used for egg hatchers and as a place for newly hatched chicks. Incubators are usually in the form of a room or box (box) of a certain size. In this research will be designed a prototype to modify the infant incubator by adding sensors to regulate the temperature and control using a Arduino microcontroller. This incubator has a measurement system and temperature regulation using LM35 sensors, heaters and fans as actuators that can cool the incubator room with Arduino as controlling. Temperature setting will be monitored by LABVIEW. In the Arduino system a program can be set up that can adjust the temperature . The temperature required by the premature baby, so that it can maintain the stability of the temperature. If the temperature in the incubator is smaller than the temperature setting, the heater will turn on and the fan speed will decrease to increase the temperature according to the settings entered. If the temperature in the incubator is greater than the temperature setting entered, the heater will turn off. The results showed that the linear regression value between the temperature values with the ADC value of y = 0.4883x with a determination coefficient of 1, while the linear regression value between the temperature with a voltage value of y = 99.481x with a coefficient of 0.9984.Keywords: Infant Incubator, temperature sensor, LM35.
RANCANG BANGUN TACHOMETER DIGITAL BERBASIS ARDUINO DILENGKAPI CHARGING DAN MODE PENYIMPAN DATA Imam Tri Harsoyo; Andi Kurniawan Nugroho; Nuriman Nuriman
Elektrika Vol. 11 No. 2 (2019): Oktober 2019
Publisher : Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (623.74 KB) | DOI: 10.26623/elektrika.v11i2.1692

Abstract

A good and feasible centrifuge is needed in the world of health, therefore a digital tachometer is needed to calibrate a centrifuge. Digital tachometer is a measuring instrument used to measure the rotation speed of a motor. This tachometer will be tested to function by being compared to using Digital Laser Photo Tachometer. This tachometer will display the number of rotations per minute of a motor according to the speed setting on the centrifuge using the output of this E18-D80NK sensor which will later be processed by the Arduino Nano. As for this feature charger and data storage mode, so that user can save the data of measurement results on SD Card, and beside that user can also be a way of charging when after usage or before usage. At testing the tool used point settings 1000, 2000, 3000, 4000, 5000, and 12000 RPM. Based on the measurement results of centrifuge using tachometer module and comparison tachometer have average percentage of error varying at each setting point. The smallest error 0.8% at the setting Point 3000 RPM, while the largest error 4.9% at the setting point 1000 RPM. The Tacometer error value on this Centrifuge measurement is still within the tolerance limit of ± 10%
Sistem Kendali Hybrid Fuzzy-Pid pada Kinematika Robot Berkaki 4 Menggunakan Sensor Gyroscope La Ode Muhamad Idris; Andi Kurniawan Nugroho; Daniyah Daniyah
Elektrika Vol. 15 No. 1 (2023): April 2023
Publisher : Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/elektrika.v15i1.3242

Abstract

Legged robots have attracted the attention of researchers because of their superior adaptation to complex environments compared to wheeled robots. Legged robots are divided into 2 (two) legged Humanoid robots, 4 (four) legged robots, 6 (six) legged robots, and other robots with more legs. Legged robots are robots that can be more adaptable to the terrain compared to wheeled robots in the case of their use in land exploration activities. Although functionally legged robots are more superior, legged robots have their own problems, namely motion control which is more complicated than wheeled robots, therefore the right method is needed to be implemented on the robot.. In this research discusses a 4 (four) legged robot designed in order to move using the inverse kinematic equation on the legs and the body of the robot which is integrated with the MPU6050 gyroscope sensor based on the Hybrid Fuzzy-PID control system. The purpose of this research is to develop a Fuzzy-PID control system that stabilizes the setpoint position in the 4 (four) legged Robot movement system. Fuzzy PID is a combination of PID control and fuzzy logic, where PID control is used to stabilize the system and fuzzy logic is used to improve the system performance. In this research, the Fuzzy-PID control system is developed using the Mamdani (Min-Max) method. The system is later tested by observing the robot's movement response to changes in the gyroscope sensor values. The results obtained were able to get an average output error up to 0.173333% during the response test to the pitch axis of -15°, but in several tests also get the response results that have a considerable error rate is up to 27.31% during the response test to the roll axis of -5°. From the test results of hybrid Fuzzy-PID control, it is obtained that the robot is able to make movements or responses to its stable point by giving reference to the x (roll), y (pitch) and z (yaw) axes where it can be analyzed that the response to the x (roll) and y (pitch) axes will affect the angle of the tibia and femur servo joints, while the response to the z (yaw) axis will affect the angle of the coxa joint servo. Keywords: IMUs (Inertial Measurement Units), Gyroscope, Inverse Kinematics, Hybrid Fuzzy-PID. ABSTRAK Robot berkaki telah menarik perhatian para peneliti karena adaptasinya unggul terhadap lingkungan yang kompleks dibanding robot beroda. Robot berkaki dibagi menjadi robot Humanoid berkaki 2(dua), robot berkaki 4(empat), robot berkaki 6(enam), dan lainnya yang berkaki lebih banyak.Robot berkaki adalah robot yang lebih adaptif terhadap medan tempuh dibandingka robot beroda dalam kasus penggunaannya pada kegiatan eksplorasi daratan. Walaupun secara fungsional robot berkaki lebih unggul, robot berkaki memiliki permasalahan sendiri, yaitu kontrol gerak yang lebih kompleks dibanding robot beroda, maka dari itu dibutuhkan metode yang tepat untuk diterapkan pada robot. Pada penelitian ini membahas tentang robot berkaki 4(empat) yang dirancang untuk dapat bergerak dengan menggunakan persamaan inverse kinematic pada kaki maupun badan robot yang diintegrasikan dengan sensor gyroscope MPU6050 berbasis sistem kendali Hybrid Fuzzy-PID.Tujuan dari penelitian ini adalah mengembangkan sistem kontrol Fuzzy-PID untuk menstabilkan posisi setpoint pada sistem pergerakan Robot berkaki 4(empat). Fuzzy PID adalah gabungan dari kontrol PID dan logika fuzzy, dimana kontrol PID digunakan untuk menstabilkan sistem dan logika fuzzy digunakan untuk memperbaiki performa sistem. Pada penelitian ini sistem kontrol Fuzzy-PID dikembangkan dengan menggunakan metode Mamdani (Min-Max). Sistem ini kemudian diuji dengan mengamati respon pergerakan robot terhadap perubahan nilai sensor gyroscope. Dimana diperoleh hasil mampu memperoleh rata-rata kesalahan hasil keluarannya sampai dengan 0.173333% pada saat pengujian respon terhadap sumbu pitch -15°, namun dibeberapa pengujian juga mendapa tkan hasil respon yang memiliki tingkat kesalahan yang cukup besar sampai dengan 27.31% pada saat pengujian respon terhadap sumbu roll -5°. Dari hasil pengujian kendali hybrid Fuzzy-PID diperoleh hasil robot mampu melakukan pergerakan atau respon untuk menuju titik stabilnya dengan pemberian acuan terhadap sumbu x(roll), y(pitch) dan z(yaw) dimana dapat dianalisa bahwa respon terhadap sumbu x(roll) dan y(pitch) akan mempengaruhi sudut servo sendi tibia dan femur, sedangkan respon terhadap sumbu z(yaw) akan mempengaruhi sudut servo sendi coxa.