Analisis Kinerja Algoritma Mesin Pembelajaran untuk Klarifikasi Penyakit Stroke Menggunakan Citra CT Scan

Stroke adalah suatu kondisi dimana pasokan darah ke otak terganggu sehingga bagian tubuh yang dikendalikan oleh area otak yang rusak tidak dapat berfungsi dengan baik. Penyebab stroke antara lain adalah terjadinya penyumbatan pada pembuluh darah (stroke iskemik) atau pecahnya pembuluh darah (stroke hemoragik). Pasien yang terkena stroke harus segera ditangani secepatnya karena sel otak dapat mati dalam hitungan menit. Tindakan penanganan stroke secara cepat dan tepat dapat mengurangi resiko kerusakan otak dan mencegah terjadinya komplikasi. Penelitian ini bertujuan untuk mengembangkan perangkat lunak yang dapat membaca dan menganalisis citra CT scan dari otak, dan kemudian secara otomatis memprediksi apakah citra CT scan tersebut stroke iskemik atau stroke hemoragik. Data citra CT scan berasal dari Rumah Sakit Umum Haji Surabaya yang diambil selama periode Januari-Mei 2019 dan berasal dari 102 pasien yang terindikasi stroke. Sebelum data gambar tersebut diolah dengan menggunakan beberapa algoritma mesin pembelajaran, data tersebut melalui tahap pre-processing yang bertujuan untuk meningkatkan kualitas citra meliputi konversi citra, pemotongan citra, penskalaan, greyscaling, penghilangan noise dan augmentasi. Tahap selanjutnya adalah ekstraksi fitur menggunakan metode Gray-Level Co-Occurrence Matrix (GLCM). Penelitian ini juga bertujuan untuk membandingkan kinerja lima algoritma mesin pembelajaran yaitu Naïve Bayes, Logistic Regression, Neural Network, Support Vector Machine dan Deep Learning yang diterapkan untuk memprediksi penyakit stroke. Hasil percobaan menunjukkan bahwa algoritma Deep Learning menghasilkan tingkat performansi paling tinggi yaitu nilai akurasi 96.78%, presisi 97.59% dan recall 95.92%. AbstractStroke is a condition in which the blood supply to the brain is interrupted so that parts of the body that are controlled by damaged brain areas cannot function properly. Causes of strokes include blockages in blood vessels (ischemic stroke) or rupture of blood vessels (hemorrhagic stroke). Stroke patients must be treated as soon as possible because brain cells can die within minutes. The handling of stroke patients quickly can reduce the risk of brain damage and prevent complications. This study aims to develop software that can read and analyze CT scan images from the brain, and then automatically predict whether the CT scan images are ischemic stroke or hemorrhagic stroke. The CT scan image data came from the Surabaya Hajj General Hospital which was taken during the January-May 2019 period and came from 102 patients who had indicated a stroke. Before the image data is processed using several machine learning algorithms, the data goes through a pre-processing phase which aims to improve image quality including image conversion, image cutting, scaling, greyscaling, noise removal and augmentation. The next step is feature extraction using the Gray-Level Co-Occurrence Matrix (GLCM) method. This study also aims to compare the performance of five machine learning algorithms, namely Naïve Bayes, Logistic Regression, Neural Networks, Support Vector Machines and Deep Learning which are applied to predict stroke. The experimental results show that the deep learning algorithm produces the highest level of performance where the accuracy value is 96.78%, 97.59% precision and 95.92% recall.

Download Full-text

Lung Cancer Detection Based On CT-Scan Images With Detection Features Using Gray Level Co-Occurrence Matrix (GLCM) and Support Vector Machine (SVM) Methods

2020 International Electronics Symposium (IES) ◽

10.1109/ies50839.2020.9231663 ◽

2020 ◽

Author(s):

Qurina Firdaus ◽

Riyanto Sigit ◽

Tri Harsono ◽

Anwar Anwar

Keyword(s):

Lung Cancer ◽

Support Vector Machine ◽

Ct Scan ◽

Cancer Detection ◽

Support Vector ◽

Gray Level ◽

Occurrence Matrix ◽

Lung Cancer Detection

Download Full-text

Hubungan Tekanan Darah dengan Volume Pendarahan Intraserebral pada Pasien Stroke Hemoragik di Ruang Rawat Inap Saraf Rumkital Dr Ramelan Surabaya

Hang Tuah Medical journal ◽

10.30649/htmj.v16i2.176 ◽

2019 ◽

Vol 16 (2) ◽

pp. 135

Author(s):

KENNY YULIAN ◽

OLIVIA MAHARDANI ADAM ◽

LESTARI DEWI

Keyword(s):

Blood Pressure ◽

Ct Scan ◽

Intracerebral Haemorrhage ◽

Hemorrhagic Stroke ◽

Haemorrhagic Stroke ◽

Cross Sectional Study ◽

Sectional Study ◽

Stroke Patients ◽

Cross Sectional ◽

The Brain

ABSTRACTBackground. Hemorrhagic stroke is a spontaneous bleeding in the brain that is usually life threatening. The most common risk factor of hemorrhagic stroke is hypertension. Hypertension can cause change in the structure of the artery wall which can cause the blood vessels near the brain to rupture easily. Purpose. To analyze the correlation between blood pressure and intracerebral haemorrhage volume in hemorrhagic stroke patients in Dr. Ramelan Navy Hospital neurologic ward. Method. This research is using a cross sectional study design. This study is done using primary data collection, using the head CT Scan result to measure the patient’s intracerebral hemorrhage volume and patient’s ER admission data for the blood pressure. Results. The study is performed to 26 haemorrhagic stroke patients who fits the inclusion and exclusion criteria. Correlation test shows no correlation between blood pressure and intracerebral haemorrhage volume in haemorrhagic stroke patients in Dr. Ramelan Navy Hospital neurologic ward, with significance (p) value of 0.888 > α (0.05). Conclusion. There is no correlation between blood pressure and intracerebral haemorrhage volume in haemorrhagic stroke patients in Dr. Ramelan Navy Hospital neurologic ward.Keywords: Blood Pressure, Intracerebral Haemorrhage Volume, Haemorrhagic Stroke ABSTRAKLatar belakang. Stroke hemoragik adalah pendarahan otak spontan yang seringkali mengancam jiwa. Faktor resiko utama terjadinya stroke hemoragik adalah adanya hipertensi. Hipertensi dapat menyebabkan perubahan struktur dinding arteri sehingga pembuluh darah didekat otak mudah ruptur. Tujuan penelitian. Mengetahui hubungan antara tekanan darah dengan volume pendarahan intraserebral pada pasien stroke hemoragik di ruang rawat inap saraf RUMKITAL Dr. Ramelan Surabaya. Metode penelitian. Penelitian ini adalah penelitian dengan desain cross sectional study. Penelitian ini dilakukan dengan perolehan data primer, yaitu berupa hasil CT scan kepala untuk mengetahui volume pendarahan intraserebral pasien dan data pasien saat masuk IGD untuk tekanan darah pasien. Hasil. Penelitian dilakukan pada 26 pasien stroke hemoragik yang masuk kriteria inklusi dan eksklusi. Hasil uji korelasi menunjukkan tidak ada hubungan antara tekanan darah dengan volume pendarahan intraserebral pada pasien stroke hemoragik di ruang rawat inap saraf RUMKITAL Dr. Ramelan Surabaya, dengan nilai signifikansi (p) = 0.888 > α (0.05). Kesimpulan. Tidak ada hubungan antara tekanan darah dengan volume pendarahan intraserebral pada pasien stroke hemoragik di ruang rawat inap saraf RUMKITAL Dr. Ramelan Surabaya.Kata Kunci : Tekanan Darah, Volume Pendarahan Intraserebral, Stroke Hemoragik

Download Full-text

Hasil pemeriksaan CT scan pada penderita stroke non hemoragik di Bagian Radiologi FK Unsrat/SMF Radiologi RSUP Prof. Dr. R. D. Kandou Manado periode Agustus 2015 – Agustus 2016

e-CliniC ◽

10.35790/ecl.4.2.2016.14398 ◽

2016 ◽

Vol 4 (2) ◽

Author(s):

Christian Elim ◽

Vonny Tubagus ◽

Ramli Hadji Ali

Keyword(s):

Ct Scan ◽

Hemorrhagic Stroke ◽

Right Hemisphere ◽

Secondary Data ◽

Haemorrhagic Stroke ◽

Descriptive Study ◽

Body Parts ◽

Stroke Patients ◽

The Right ◽

The Brain

Abstract: CT-scan is used to analyze the structures of specific body parts, mainly to confirm the diagnosis of non-hemorrhagic stroke. Stroke is a neurological deficit that occurs suddenly and caused by the interruption of blood flow to the brain. The symptoms are corresponding to the location of the stroke. This study was aimed to obtain the CT-scan examination of non-haemorrhagic stroke patients. This was a retrospective descriptive study using secondary data such as request letter and data of head CT-scan performed from August 2015 to August 2016. The results showed that there were 89 cases of non-hemorrhagic stroke. The majority were males (60 patients; 67%), elderly ≥65 years old (27 patients; 30%), and location of lesion in the right hemisphere (38 patients; 43%). Conclusion: In this study most patients diagnosed as non-hemorrhagic stroke with CT-scan were males, over 65 years old, and location of lesion in right hemisphere.Keywords: non-haemorrhagic stroke, CT-scan Abstrak: CT scan digunakan untuk menganalisis struktur dalam dari beberapa bagian tubuh tertentu, antara lain untuk memastikan diagnosis dari stroke non hemoragik, Stroke merupakan suatu defisit neurologik yang terjadi secara tiba-tiba diakibatkan oleh adanya gangguan aliran darah ke otak dan gejala yang terjadi sesuai dengan lokasi dari stroke tersebut. Penelitian ini bertujuan untuk mengetahui hasil pemeriksaan CT scan pada penderita stroke non hemoragik. Jenis penelitian ialah deskriptif retrospektif dengan memanfaatkan data sekunder berupa lembar permintaan dan data hasil CT scan kepala yang dilaksanakan sejak Agustus 2015 sampai Agustus 2016. Hasil penelitian mendapatkan sebanyak 89 kasus didiagnosis stroke non-hemoragik dengan CT-scan, terbanyak ialah jenis kelamin laki-laki berjumlah 60 orang (67%); golongan usia manula (≥65 tahun) berjumlah 27 orang (30%); dan lokasi lesi di hemisfer dekstra berjumlah 38 orang (43%). Simpulan: Pada studi ini, majoritas pasien yang didiagnosis stroke non-hemoragik dengan CT-scan Berjenis kelamin laki-laki, usia ≥65 tahun, dengan lokais lesi pada hemisfer kanan. Kata kunci: stroke non hemoragik, CT-scan

Download Full-text

GAMBARAN HASIL PEMERIKSAAN CT SCAN KEPALA PADA PENDERITA STROKE HEMORAGIK DI BAGIAN RADIOLOGI FK. UNSRAT/SMF RADIOLOGI BLU RSUP PROF. DR. R. D. KANDOU MANADO

e-CliniC ◽

10.35790/ecl.3.1.2015.6746 ◽

2015 ◽

Vol 3 (1) ◽

Author(s):

Novita T. A. Parinding ◽

Ramli Haji Ali ◽

Vonny N. Tubagus

Keyword(s):

Ct Scan ◽

Hemorrhagic Stroke ◽

Recurrent Stroke ◽

Secondary Data ◽

Chronic Hypertension ◽

Inclusion Criteria ◽

Cerebral Blood Vessels ◽

Stroke Patients ◽

Head Ct ◽

The Brain

Abstract: Hemorrhagic stroke is a spontaneous blood vessel bursts inside the brain. The main cause is chronic hypertension and the presence of degeneration of cerebral blood vessels. Bleeding can occur in the brain and the subarachnoid space due to rupture of an artery or rupture of the aneurysm. At a stroke, a CT scan is the gold standard examination to distinguish infarction with bleeding, because CT scan can give a very clear picture of the head of the intracranial space persisted as cerebral infarction, intracranial hemorrhage, and hemorrhagic stroke. So it can be helpful to diagnose the disease and neurological disorders. The purpose of this study is to cognize the distribution of hemorrhagic stroke patients who performed a head CT scan at The Department of Radiology of FK UNSRAT / SMF Radiology of BLU Prof. Dr. R. D. Kandou Manado. This is a retrospective descriptive study by using secondary data coming from hemorragic stroke patients medical records in the Department of Radiology of BLU Prof. Dr. R. D. Kandou Manado on the period time from May to October 2014. Patients admitted inclusion criteria age between 30-80 years old, proven hemorrhagic stroke based on history, physical examination and investigations of doctors. The results showed that the total number of head CT Scans are 296 patients and more showed abnormal picture (58.1%) compared with the normal picture (41.9%), there are 64 patients with abnormal CT Scan picture results of hemorrhagic stroke (37.2% ), patients with hemorrhagic stroke are more common in males (65.6%), most in the age group of the early elderly betweeb 46-55 years old (40.6%) with most bleeding type of intracerebral hemorrhage (87.5%). Conclusion: Patient that comes with hemorrhagic stroke case or recurrent stroke should undertake a CT Scan of head to assist diagnose and later on can identify the type of bleeding caused by the stroke.Keywords: head CT scan, haemorrhagic strokeAbstrak: Stroke hemoragik adalah perdarahan spontan di dalam otak. Penyebab utamanya adalah hipertensi kronik dan adanya degenerasi pembuluh darah cerebral. Perdarahan dapat terjadi di dalam otak dan ruang subaraknoid karena ruptur dari arteri atau ruptur dari aneurisma. Pada penyakit stroke, CT Scan merupakan pemeriksaan baku emas untuk membedakan infark dengan perdarahan, karena CT Scan dapat memberikan gambaran kepala yang sangat jelas tentang proses desak ruang intrakranial seperti infark otak, perdarahan intrakranial, dan stroke hemoragik. Sehingga dapat membantu penegakan diagnosis penyakit dan kelainan neurologik. Tujuan penelitian ini untuk mengetahui distribusi penderita stroke hemoragik yang dilakukan pemeriksaan CT Scan kepala di Bagian Radiologi FK UNSRAT/SMF Radiologi BLU RSUP Prof. Dr. R. D. Kandou Manado. Penelitian ini merupakan penelitian deskriptif retrospektif dengan memanfaatkan data sekunder berupa rekam medik pasien stroke hemoragik di Bagian Radiologi BLU RSUP Prof. Dr. R. D. Kandou Manado periode Mei-Oktober 2014. Pasien yang masuk kriteria inklusi yaitu usia 30-80 tahun, terbukti stroke hemoragik berdasarkan anamnesis, pemeriksaan fisik dan penunjang oleh dokter. Hasil pemeriksaan CT Scan kepala berjumlah 296 pasien dan lebih banyak menunjukkan gambaran abnormal (58,1%) dibandingkan gambaran normal (41,9%), pada gambaran abnormal terdapat 64 penderita dengan hasil CT Scan gambaran stroke hemoragik (37,2%), penderita stroke hemoragik lebih banyak terjadi pada laki-laki (65,6%), paling banyak pada kelompok umur lansia awal 46-55 tahun (40,6%) dengan tipe perdarahan paling banyak yaitu perdarahan intraserebral (87,5%). Simpulan : Penderita yang datang dengan keluhan stroke hemoragik atau stroke berulang sebaiknya melakukan pemeriksaan CT Scan kepala untuk membantu diagnosis dan dapat diketahui tipe perdarahan dari stroke tersebut.Kata kunci : CT scan kepala, stroke hemoragik

Download Full-text

GAMBARAN HASIL CT SCAN KEPALA PADA PENDERITA DENGAN KLINIS STROKE NON-HEMORAGIK DI BAGIAN RADIOLOGI FK. UNSRAT / SMF RADIOLOGI BLU RSUP PROF. DR. R. D KANDOU MANADO PERIODE JANUARI 2011- DESEMBER 2011

e-CliniC ◽

10.35790/ecl.2.3.2014.6011 ◽

2014 ◽

Vol 2 (3) ◽

Author(s):

Mohammad Arswendo Tjikoe ◽

Elvie Loho ◽

Ramli H. Ali

Keyword(s):

Blood Flow ◽

Ischemic Stroke ◽

Ct Scan ◽

Hemorrhagic Stroke ◽

Age Group ◽

Stroke Patients ◽

Request Form ◽

Neurologic Diseases ◽

Short Time ◽

The Brain

Abstract: Stroke is the most common of neurologic manifestations and easily recognizable from the other neurologic diseases due to the early onset of sudden in a short time. Stroke as clinical diagnosis was divided to hemorrhagic stroke and ischemic stroke. In hemorrhagic stroke there is a rupture in blood vessel so the blood flow became abnormal and bleeds into surrounding brain and damage it. In ischemic stroke the blood flow heading to the brain is interrupted due to atherosclerosis process. The purpose of this study is to know about description of head CT scan in patient with clinical diagnonis of stroke non hemorrhagic in Department/SMF Radiology Faculty Of Medicine UNSRAT BLU RSUP Prof. Dr. R. D. Kandou Manado period on 1st January 2011 – 31st December 2011. Methods: The study design was a retrospective descriptive study. The data are from request form sheet and radiographic response in the Department of Radiology and processed in descriptive. Results: Base on 163 data of stroke patients obtained, 74 patients diagnosed with infarction stroke (45,4%). Male had more (59,5%) than female (40,5%). For age group, 60-79 is the largest with 33 patients (44,6%). Area with most lesion was in parietal dextra lobe with 8 cases (10,8%). Most cases was happened in August with 10 cases (13,5%). Conclusion: Patients with radiology diagnosis infarction stroke, the most common infarction location is in parietal dextra area. Keywords: CT Scan, Infarction Stroke, Parietal Dextra. Abstrak: Stroke merupakan salah satu manifestasi neurologik yang umum, dan mudah dikenal dari penyakit-penyakit neurologik lain karena mula timbulnya mendadak dalam waktu yang singkat. Stroke sebagai diagnosis klinis terbagi menjadi stroke hemoragik (pendarahan) dan stroke non-hemoragik (iskemik). Pada stroke hemoragik pembuluh darah pecah sehingga aliran darah menjadi tidak normal dan darah yang keluar merembes masuk ke dalam suatu daerah di otak dan merusaknya. Sedangkan pada stroke non-hemoragik aliran darah ke otak terhenti karena aterosklerosis atau bekuan darah yang telah menyumbat suatu pembuluh darah, melalui proses aterosklerosis. Tujuan penelitian ini adalah untuk mengetahui gambaran hasil CT scan kepala pada penderita dengan klinis stroke non-hemoragik di Bagian Radiologi FK. Unsrat / SMF Radiologi BLU RSUP Prof. dr. R. D Kandou Manado periode Januari 2011- Desember 2011. Metode: Penelitian ini merupakan penelitian deskriptif retrospektif dengan memanfaatkan data sekunder berupa lembaran permintaan & jawaban CT scan kepala yang terdapat di bagian Radiologi BLU RSUP Prof. Dr. R. D. Kandou Manado periode 1 Januari 2011 – 31 Desember 2011. Hasil penelitian: Berdasarkan 163 data pasien yang didapatkan, 74 pasien didiagnosis dengan stroke infark (45,4%). Laki-laki lebih banyak (59,5%) dari perempuan (40,5%). Kelompok umur 60-79 merupakan kelompok umur terbanyak yaitu 33 pasien (44,6%). Daerah lesi terbanyak adalah pada daerah parietalis dextra dengan 8 kasus (10,8%). Kasus terbanyak terjadi pada bulan agustus dengan 10 kasus (13,5%). Simpulan: Pada pasien dengan diagnosis radiologi stroke infark, lokasi infark yang paling banyak muncul adalah terdapat pada daerah parietal dextra. Kata kunci: CT Scan, Stroke Infark, Parietal Dextra.

Download Full-text

Implementation of IoT Framework with Data Analysis Using Deep Learning Methods for Occupancy Prediction in a Building

Future Internet ◽

10.3390/fi13030067 ◽

2021 ◽

Vol 13 (3) ◽

pp. 67

Author(s):

Eric Hitimana ◽

Gaurav Bajpai ◽

Richard Musabe ◽

Louis Sibomana ◽

Jayavel Kayalvizhi

Keyword(s):

Machine Learning ◽

Time Series ◽

Deep Learning ◽

Time Series Data ◽

Multivariate Time Series ◽

Machine Learning Algorithms ◽

Series Data ◽

Support Vector ◽

Human Beings ◽

Feed Forward Network

Many countries worldwide face challenges in controlling building incidence prevention measures for fire disasters. The most critical issues are the localization, identification, detection of the room occupant. Internet of Things (IoT) along with machine learning proved the increase of the smartness of the building by providing real-time data acquisition using sensors and actuators for prediction mechanisms. This paper proposes the implementation of an IoT framework to capture indoor environmental parameters for occupancy multivariate time-series data. The application of the Long Short Term Memory (LSTM) Deep Learning algorithm is used to infer the knowledge of the presence of human beings. An experiment is conducted in an office room using multivariate time-series as predictors in the regression forecasting problem. The results obtained demonstrate that with the developed system it is possible to obtain, process, and store environmental information. The information collected was applied to the LSTM algorithm and compared with other machine learning algorithms. The compared algorithms are Support Vector Machine, Naïve Bayes Network, and Multilayer Perceptron Feed-Forward Network. The outcomes based on the parametric calibrations demonstrate that LSTM performs better in the context of the proposed application.

Download Full-text

A semiautomatic annotation approach for sentiment analysis

Journal of Information Science ◽

10.1177/01655515211006594 ◽

2021 ◽

pp. 016555152110065

Author(s):

Rahma Alahmary ◽

Hmood Al-Dossari

Keyword(s):

Deep Learning ◽

Sentiment Analysis ◽

Short Term Memory ◽

Machine Learning Algorithms ◽

Support Vector ◽

Short Term ◽

Term Memory ◽

Annotation Process ◽

Learning Classifiers ◽

Long Short Term Memory

Sentiment analysis (SA) aims to extract users’ opinions automatically from their posts and comments. Almost all prior works have used machine learning algorithms. Recently, SA research has shown promising performance in using the deep learning approach. However, deep learning is greedy and requires large datasets to learn, so it takes more time for data annotation. In this research, we proposed a semiautomatic approach using Naïve Bayes (NB) to annotate a new dataset in order to reduce the human effort and time spent on the annotation process. We created a dataset for the purpose of training and testing the classifier by collecting Saudi dialect tweets. The dataset produced from the semiautomatic model was then used to train and test deep learning classifiers to perform Saudi dialect SA. The accuracy achieved by the NB classifier was 83%. The trained semiautomatic model was used to annotate the new dataset before it was fed into the deep learning classifiers. The three deep learning classifiers tested in this research were convolutional neural network (CNN), long short-term memory (LSTM) and bidirectional long short-term memory (Bi-LSTM). Support vector machine (SVM) was used as the baseline for comparison. Overall, the performance of the deep learning classifiers exceeded that of SVM. The results showed that CNN reported the highest performance. On one hand, the performance of Bi-LSTM was higher than that of LSTM and SVM, and, on the other hand, the performance of LSTM was higher than that of SVM. The proposed semiautomatic annotation approach is usable and promising to increase speed and save time and effort in the annotation process.

Download Full-text

A Comparative Study of Alzheimer Detection Techniques

International Journal for Research in Applied Science and Engineering Technology ◽

10.22214/ijraset.2021.37882 ◽

2021 ◽

Vol 9 (8) ◽

pp. 2877-2883

Author(s):

Adwait Patil

Keyword(s):

Machine Learning ◽

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Deep Learning ◽

Comparative Study ◽

Fuzzy System ◽

Early Stage ◽

Machine Learning Algorithms ◽

Support Vector ◽

Detection Techniques

Abstract: Alzheimer’s disease is one of the neurodegenerative disorders. It initially starts with innocuous symptoms but gradually becomes severe. This disease is so dangerous because there is no treatment, the disease is detected but typically at a later stage. So it is important to detect Alzheimer at an early stage to counter the disease and for a probable recovery for the patient. There are various approaches currently used to detect symptoms of Alzheimer’s disease (AD) at an early stage. The fuzzy system approach is not widely used as it heavily depends on expert knowledge but is quite efficient in detecting AD as it provides a mathematical foundation for interpreting the human cognitive processes. Another more accurate and widely accepted approach is the machine learning detection of AD stages which uses machine learning algorithms like Support Vector Machines (SVMs) , Decision Tree , Random Forests to detect the stage depending on the data provided. The final approach is the Deep Learning approach using multi-modal data that combines image , genetic data and patient data using deep models and then uses the concatenated data to detect the AD stage more efficiently; this method is obscure as it requires huge volumes of data. This paper elaborates on all the three approaches and provides a comparative study about them and which method is more efficient for AD detection. Keywords: Alzheimer’s Disease (AD), Fuzzy System , Machine Learning , Deep Learning , Multimodal data

Download Full-text

Automatic Grading of Stroke Symptoms for Rapid Assessment Using Optimized Machine Learning and 4-Limb Kinematics: Clinical Validation Study (Preprint)

10.2196/preprints.20641 ◽

2020 ◽

Author(s):

Eunjeong Park ◽

Kijeong Lee ◽

Taehwa Han ◽

Hyo Suk Nam

Keyword(s):

Machine Learning ◽

Medical Staff ◽

Machine Learning Algorithms ◽

Training Data ◽

Support Vector ◽

Grading System ◽

Operating Characteristics ◽

Stroke Patients ◽

Motor Weakness ◽

Automatic Grading

BACKGROUND Subtle abnormal motor signs are indications of serious neurological diseases. Although neurological deficits require fast initiation of treatment in a restricted time, it is difficult for nonspecialists to detect and objectively assess the symptoms. In the clinical environment, diagnoses and decisions are based on clinical grading methods, including the National Institutes of Health Stroke Scale (NIHSS) score or the Medical Research Council (MRC) score, which have been used to measure motor weakness. Objective grading in various environments is necessitated for consistent agreement among patients, caregivers, paramedics, and medical staff to facilitate rapid diagnoses and dispatches to appropriate medical centers. OBJECTIVE In this study, we aimed to develop an autonomous grading system for stroke patients. We investigated the feasibility of our new system to assess motor weakness and grade NIHSS and MRC scores of 4 limbs, similar to the clinical examinations performed by medical staff. METHODS We implemented an automatic grading system composed of a measuring unit with wearable sensors and a grading unit with optimized machine learning. Inertial sensors were attached to measure subtle weaknesses caused by paralysis of upper and lower limbs. We collected 60 instances of data with kinematic features of motor disorders from neurological examination and demographic information of stroke patients with NIHSS 0 or 1 and MRC 7, 8, or 9 grades in a stroke unit. Training data with 240 instances were generated using a synthetic minority oversampling technique to complement the imbalanced number of data between classes and low number of training data. We trained 2 representative machine learning algorithms, an ensemble and a support vector machine (SVM), to implement auto-NIHSS and auto-MRC grading. The optimized algorithms performed a 5-fold cross-validation and were searched by Bayes optimization in 30 trials. The trained model was tested with the 60 original hold-out instances for performance evaluation in accuracy, sensitivity, specificity, and area under the receiver operating characteristics curve (AUC). RESULTS The proposed system can grade NIHSS scores with an accuracy of 83.3% and an AUC of 0.912 using an optimized ensemble algorithm, and it can grade with an accuracy of 80.0% and an AUC of 0.860 using an optimized SVM algorithm. The auto-MRC grading achieved an accuracy of 76.7% and a mean AUC of 0.870 in SVM classification and an accuracy of 78.3% and a mean AUC of 0.877 in ensemble classification. CONCLUSIONS The automatic grading system quantifies proximal weakness in real time and assesses symptoms through automatic grading. The pilot outcomes demonstrated the feasibility of remote monitoring of motor weakness caused by stroke. The system can facilitate consistent grading with instant assessment and expedite dispatches to appropriate hospitals and treatment initiation by sharing auto-MRC and auto-NIHSS scores between prehospital and hospital responses as an objective observation.

Download Full-text

Lung cancer classification using fuzzy c-means and fuzzy kernel C-Means based on CT scan image

IAES International Journal of Artificial Intelligence (IJ-AI) ◽

10.11591/ijai.v10.i2.pp291-297 ◽

2021 ◽

Vol 10 (2) ◽

pp. 291

Author(s):

Zuherman Rustam ◽

Aldi Purwanto ◽

Sri Hartini ◽

Glori Stephani Saragih

Keyword(s):

Lung Cancer ◽

Ct Scan ◽

Lung Nodule ◽

Image Data ◽

Cluster Center ◽

Support Vector ◽

Fuzzy C Means ◽

Kernel Distance ◽

Abnormal Cells ◽

Membership Matrix

Cancer is one of the diseases with the highest mortality rate in the world. Cancer is a disease when abnormal cells grow out of control that can attack the body's organs side by side or spread to other organs. Lung cancer is a condition when malignant cells form in the lungs. To diagnose lung cancer can be done by taking x-ray images, CT scans, and lung tissue biopsy. In this modern era, technology is expected to help research in the field of health. Therefore, in this study feature extraction from CT images was used as data to classify lung cancer. We used CT scan image data from SPIE-AAPM Lung CT challenge 2015. Fuzzy C-Means and fuzzy kernel C-Means were used to classify the lung nodule from the patient into benign or malignant. Fuzzy C-Means is a soft clustering method that uses Euclidean distance to calculate the cluster center and membership matrix. Whereas fuzzy kernel C-Means uses kernel distance to calculate it. In addition, the support vector machine was used in another study to obtain 72% average AUC. Simulations were performed using different k-folds. The score showed fuzzy kernel C-Means had the highest accuracy of 74%, while fuzzy C-Means obtained 73% accuracy.

Download Full-text