Computer classification of experimental brain tumors in mice

1988 ◽  
Vol 35 (1) ◽  
pp. 41-46 ◽  
Author(s):  
H. Kroh ◽  
J.R. Iglesias ◽  
E. Matyja ◽  
C. Aruffo ◽  
K. Meier-Hauf ◽  
...  
Author(s):  
Saleh Alaraimi ◽  
Kenneth E. Okedu ◽  
Hugo Tianfield ◽  
Richard Holden ◽  
Omair Uthmani

2021 ◽  
Vol 11 (3) ◽  
pp. 352
Author(s):  
Isselmou Abd El Kader ◽  
Guizhi Xu ◽  
Zhang Shuai ◽  
Sani Saminu ◽  
Imran Javaid ◽  
...  

The classification of brain tumors is a difficult task in the field of medical image analysis. Improving algorithms and machine learning technology helps radiologists to easily diagnose the tumor without surgical intervention. In recent years, deep learning techniques have made excellent progress in the field of medical image processing and analysis. However, there are many difficulties in classifying brain tumors using magnetic resonance imaging; first, the difficulty of brain structure and the intertwining of tissues in it; and secondly, the difficulty of classifying brain tumors due to the high density nature of the brain. We propose a differential deep convolutional neural network model (differential deep-CNN) to classify different types of brain tumor, including abnormal and normal magnetic resonance (MR) images. Using differential operators in the differential deep-CNN architecture, we derived the additional differential feature maps in the original CNN feature maps. The derivation process led to an improvement in the performance of the proposed approach in accordance with the results of the evaluation parameters used. The advantage of the differential deep-CNN model is an analysis of a pixel directional pattern of images using contrast calculations and its high ability to classify a large database of images with high accuracy and without technical problems. Therefore, the proposed approach gives an excellent overall performance. To test and train the performance of this model, we used a dataset consisting of 25,000 brain magnetic resonance imaging (MRI) images, which includes abnormal and normal images. The experimental results showed that the proposed model achieved an accuracy of 99.25%. This study demonstrates that the proposed differential deep-CNN model can be used to facilitate the automatic classification of brain tumors.


2017 ◽  
Vol 19 (suppl_6) ◽  
pp. vi181-vi181
Author(s):  
Quin Xie ◽  
Dominick Han ◽  
Kevin Faust ◽  
Kenneth Aldape ◽  
Gelareh Zadeh ◽  
...  

1997 ◽  
Vol 17 (11) ◽  
pp. 1239-1253 ◽  
Author(s):  
Hisao Uehara ◽  
Tadashi Miyagawa ◽  
Juri Tjuvajev ◽  
Revathi Joshi ◽  
Bradley Beattie ◽  
...  

The goal of this study was to evaluate the differences and define the advantages of imaging experimental brain tumors in rats with two nonmetabolized amino acids, 1-aminocyclopentane carboxylic (ACPC) acid and α-aminoisobutyric (AIB) acid compared with imaging with fluorodeoxyglucose (FDG) or the gallium-diethylenetriaminepentaacetic acid chelate (Ga-DTPA). 1-aminocyclopentane carboxylic acid, AIB, and FDG autoradiograms were obtained 60 minutes after intravenous injection to simulate positron emission tomography (PET) imaging, whereas the Ga-DTPA autoradiograms were obtained 5 or 10 minutes after injection to simulate gadolinium (Gd)-DTPA–enhanced magnetic resonance (MR) images. Three experimental tumors were studied (C6, RG2, and Walker 256) to provide a range of tumor types. Triple-label quantitative autoradiography was performed, and parametric images of the apparent distribution volume (Va, mL/g) for ACPC or AIB, relative glucose metabolism (R, μmol/100 g/min), vascular permeability to Ga-DTPA (K1, μL/min/g), and histology were obtained from the same tissue section. The four images were registered in an image array processor, and regions of interest in tumor and contralateral brain were defined on morphologic criteria (histology) and were transferred to the autoradiographic images. A comparative analysis of all measured values was performed. The location and morphologic characteristics of the tumor had an effect on the images and measurements of Va, R, and K1. Meningeal extensions of all three tumors consistently had the highest amino acid uptake (Va) and vascular permeability (K1) values, and subcortical portions of the tumors usually had the lowest values. Va and R (FDG) values generally were higher in tumor regions with high-cell density and lower in regions with low-cell density. Tumor areas identified as “impending” necrosis on morphologic criteria consistently had high R values, but little or no change in Va or K1. Tumor necrosis was seen consistently only in the larger Walker 256 tumors; low values of R and Va for AIB (less for ACPC) were measured in the necrotic-appearing regions, whereas K1 was not different from the mean tumor value. The highest correlations were observed between vascular permeability (K1 for Ga-DTPA) and Va for AIB in all three tumors; little or no correlation between vascular permeability and R was observed. The advantages of ACPC and AIB imaging were most convincingly demonstrated in C6 gliomas and in Walker 256 tumors. 1-aminocyclopentane was substantially better than FDG or Ga-DTPA for identifying tumor infiltration of adjacent brain tissue beyond the macroscopic border of the tumor; ACPC also may be useful for identifying low-grade tumors with an intact blood–brain barrier. Contrast-enhancing regions of the tumors were visualized more clearly with AIB than with FDG or Ga-DTPA; viable and necrotic-appearing tumor regions could be distinguished more readily with AIB than with FDG. [11C]-labeled ACPC and AIB are likely to have similar advantages for imaging human brain tumors with PET.


1983 ◽  
Vol 23 (2) ◽  
pp. 109-115
Author(s):  
Katsuzo KIYA ◽  
Hirofumi OKAMOTO ◽  
Kiyoshi HARADA ◽  
Tohru UOZUMI ◽  
Tetsuya TOGE ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document