Isi Piringku Dietary Meal Proportion Estimator Applications using SeeFood Image Segmentations

Isi Piringku atau My Meal Dish Content was a program initiated by the Indonesian ministry of health to promote a healthy daily lifestyle consisting of balanced dietary, enough hydration, active lifestyle, cleanliness and body weight control. The balanced diet meal was supposed to consist of ⅓ of carbohydrate intake, ⅓ of vegetable intake, ⅙ of fruit intake and ⅙ of protein intake every time. This introduces some difficulty that every meal must be measured to align with the dietary guidelines. This study targets estimating the meal diet proportion by its visual cues using smartphone application. While the actual meal content dietary division was weight based, for sake of simplicity the proportion in this study was estimated by each food area which roughly correlates to its volume. Using smartphone cameras in Android 9 Operating Systems and Tensorflow Lite Seefoods: Mobile Food Segmentation v1.0 module, an application was built to help users estimate their meal balances proportion. The original segmentation criterion was constructed using USDA dietary guidelines and it was reduced to only 4 food groups related to Isi Piringku criterion. Suggestion will be given regarding the segmentation result. The result was that the application was capable of estimating the meal diet proportion and giving suggestions based on the segmentation result. Although, the volume of the meal food groups estimated was still low on accuracy. This was correlated with the accuracy level of the segmentation module that was used. On average, the time needed to apply the segmentation process was around 2 to 3 seconds on a Snapdragon 835 device.

Development of segmentation algorithm for determining planktonic objects from microscopic images

Plankton are free-floating organisms that live, grow, and move along with the ocean currents. This free-floating organism plays important roles as primary producers, they serve as a link to energy transfer, and a factor that regulates the biogeochemical cycles. Indonesia, with almost 60% of its territory covered by the ocean, harbours a wide variety of planktonic species. However, one of the issues within usual planktonic studies is the lack of a fast and accurate method for identifying and classifying the plankton type. Thus, the computer vision methods on microscopic images were proposed to deal with the problem. The classification follows two main steps, detecting plankton location and followed by plankton differentiation. The segmentation algorithm is required to limit the determination area. The present study describes the segmentation methods on fifteen plankton types. The U-Net based architecture was implemented to segment the plankton texture from other objects. The segmentation result was also compared with the manual assessment to compute the performance parameters. The accuracy, 0.970±0.025, gives the highest value whereas the smallest value is found in the precision parameter, 0.761±0.156.

Remote Sensing Image Semantic Segmentation Algorithm Based on Improved ENet Network

A remote sensing image semantic segmentation algorithm based on improved ENet network is proposed to improve the accuracy of segmentation. First, dilated convolution and decomposition convolution are introduced in the coding stage. They are used in conjunction with ordinary convolution to increase the receptive field of the model. Each convolution output contains a larger range of image information. Second, in the decoding stage, the image information of different scales is obtained through the upsampling operation and then through the compression, excitation, and reweighting operations of the Squeeze and Excitation (SE) module. The weight of each feature channel is recalibrated to improve the accuracy of the network. Finally, the Softmax activation function and the Argmax function are used to obtain the final segmentation result. Experiments show that our algorithm can significantly improve the accuracy of remote sensing image semantic segmentation.

Performance Analysis on the Effect of Noise in Inverse Surface Adaptive Thresholding (ISAT)

Thresholding is one of the powerful methods in segmentation phase. Numerous methods were proposed to segment the foreground from the background but there is limited number of studies that analyse the effect of noise since the present of noise will affect the performance of the thresholding method. In this paper, the main idea is to analyse the effect of noise in Inverse Surface Adaptive Thresholding (ISAT) method. ISAT method is known as an excellent method to segment the image with the present of noise. The result of this analysis can be a guideline to researcher when implementing ISAT method especially in medical image diagnosis. Initially, several images with different noise variations were prepared and underwent ISAT method. In ISAT method, several image processing methods were incorporated namely edge detection, Otsu thresholding and inverse surface construction. The resulting images were evaluated using Misclassification Error (ME) to evaluate the performance of the segmentation result. Based on the obtained results, ISAT performance is consistent although the noise percentage increases from 5% to 25%.

Detection of 3D Arterial Centerline Extraction in Spiral CT Coronary Angiography

This paper presents an in-depth study and analysis of the 3D arterial centerline in spiral CT coronary angiography, and constructs its detection and extraction technique. The first time, the distance transform is used to complete the boundary search of the original figure; the second time, the distance transform is used to calculate the value of the distance transform of all voxels, and according to the value of the distance transform, unnecessary voxels are deleted, to complete the initial contraction of the vascular region and reduce the computational consumption in the next process; then, the nonwitnessed voxels are used to construct the maximum inner joint sphere model and find the skeletal voxels that can reflect the shape of the original figure. Finally, the skeletal lines were optimized on these initially extracted skeletal voxels using a dichotomous-like principle to obtain the final coronary artery centerline. Through the evaluation of the experimental results, the algorithm can extract the coronary centerline more accurately. In this paper, the segmentation method is evaluated on the test set data by two kinds of indexes: one is the index of segmentation result evaluation, including dice coefficient, accuracy, specificity, and sensitivity; the other is the index of clinical diagnosis result evaluation, which is to refine the segmentation result for vessel diameter detection. The results obtained in this paper were compared with the physicians’ labeling results. In terms of network performance, the Dice coefficient obtained in this paper was 0.89, the accuracy was 98.36%, the sensitivity was 93.36%, and the specificity was 98.76%, which reflected certain advantages in comparison with the advanced methods proposed by previous authors. In terms of clinical evaluation indexes, by performing skeleton line extraction and diameter calculation on the results obtained by the segmentation method proposed in this paper, the absolute error obtained after comparing with the diameter of the labeled image was 0.382 and the relative error was 0.112, which indicates that the segmentation method in this paper can recover the vessel contour more accurately. Then, the results of coronary artery centerline extraction with and without fine branch elimination were evaluated, which proved that the coronary artery centerline has higher accuracy after fine branch elimination. The algorithm is also used to extract the centerline of the complete coronary artery tree, and the results prove that the algorithm has better results for the centerline extraction of the complete coronary vascular tree.

A Dense RNN for Sequential Four-Chamber View Left Ventricle Wall Segmentation and Cardiac State Estimation

The segmentation of the left ventricle (LV) wall in four-chamber view cardiac sequential image is significant for cardiac disease diagnosis and cardiac mechanisms study; however, there is no successful reported work on sequential four-chambered view LV wall segmentation due to the complex four-chamber structure and diversity of wall motion. In this article, we propose a dense recurrent neural network (RNN) algorithm to achieve accurately LV wall segmentation in a four-chamber view MRI time sequence. In the cardiac sequential LV wall process, not only the sequential accuracy but also the accuracy of each image matters. Thus, we propose a dense RNN to provide compensation for the first long short-term memory (LSTM) cells. Two RNNs are combined in this work, the first one aims at providing information for the first image, and the second RNN generates segmentation result. In this way, the proposed dense RNN improves the accuracy of the first frame image. What is more is that, it improves the effectiveness of information flow between LSTM cells. Obtaining more competent information from the former cell, frame-wise segmentation accuracy is greatly improved. Based on the segmentation result, an algorithm is proposed to estimate cardiac state. This is the first time that deals with both cardiac time-sequential LV segmentation problems and, robustly, estimates cardiac state. Rather than segmenting each frame separately, utilizing cardiac sequence information is more stable. The proposed method ensures an Intersection over Union (IoU) of 92.13%, which outperforms other classical deep learning algorithms.

Sequential U-Net Architecture for Automatic Femoral Artery Segmentation in Ultrasound Images

This work compares three different approaches to automatically segment the femoral artery from 2D ultrasound images. Two of the architectures follow a sequential structure, where each ultrasound image is considered a slice of the whole vessel volume, and its previous segmentation result will be part of the input, thus leading to a spatial prior. The Dice score on test data show a better performance on the baseline U-Net (0.819) compared to the sequential U-Net approaches (0.633, 0.725) for the femoral artery segmentation. This could be attributed to the misalignment of the slices being used in those networks. A possible improvement could be assumed in the implementation of a spatially calibrated and tracked ultrasound probe. Overall, these results indicate promising approaches for an automatic segmentation of the femoral artery using 2D ultrasound data.

Adoption of Snake Variable Model-Based Method in Segmentation and Quantitative Calculation of Cardiac Ultrasound Medical Images

This paper intends to explore the effect of the enhanced snake variable model in the segmentation of cardiac ultrasound images and its adoption in quantitative measurement of cardiac cavity. First, the basic principles of the traditional snake model and the gradient vector flow (GVF) snake model are explained. Then, an ellipsoid model is constructed to obtain the initial contour of the heart based on the three-dimensional volume of cardiac ultrasound image, and a discretized triangular mesh model is generated. Finally, the vortical gradient vector flow (VGVF) external force field is introduced and combined with the greedy algorithm to process the deformation of the initial ellipsoid contour of the heart. The segmentation effect is quantitatively evaluated regarding the area overlap rate (AOR) and the mean contour distance (MCD). The results show that the VGVF snake model can segment the deep recessed area of the “U-shaped map” in contrast to the traditional snake model and the GVF snake model. After being applied to ultrasonic image segmentation, the VGVF snake model obtains the segmentation result that is close to the doctor’s manual segmentation result, and the average AOR and MCD are 97.4% and 3.2, respectively. The quantitative evaluation of the cardiac cavity is carried out based on the segmentation results, and the measurement of the volume change of the left ventricle within a cardiac cycle is realized. To sum up, VGVF snake model is superior to the traditional snake and GVF snake models in terms of ultrasonic image segmentation, which realizes the three-dimensional segmentation and quantitative calculation of the cardiac cavity.

Cell image segmentation by using feedback and convolutional LSTM

Human brain is known to have a layered structure and perform not only feedforward process from lower layer to upper layer, but also feedback process from upper layer to lower layer. Neural network is a mathematical model of the function of neurons, and several models are proposed until now. Although neural network imitates the human brain, everyone uses only feedforward process and direct feedback process from upper layer to lower layer is not used in prediction process. Therefore, in this paper, we propose Feedback U-Net using convolutional LSTM. Our model is a segmentation model using convolutional LSTM and feedback process. The output of U-Net at the first round is fed back to the input, and our method re-considers the segmentation result at the second round. By using convolutional LSTM, the features are extracted well based on the features extracted at the first round. On both of the Drosophila cell image and Mouse cell image datasets, our model outperformed conventional U-Net which uses only feedforward process.

Multi-Resolution Supervision Network with an Adaptive Weighted Loss for Desert Segmentation

Desert segmentation of remote sensing images is the basis of analysis of desert area. Desert images are usually characterized by large image size, large-scale change, and irregular location distribution of surface objects. The multi-scale fusion method is widely used in the existing deep learning segmentation models to solve the above problems. Based on the idea of multi-scale feature extraction, this paper took the segmentation results of each scale as an independent optimization task and proposed a multi-resolution supervision network (MrsSeg) to further improve the desert segmentation result. Due to the different optimization difficulty of each branch task, we also proposed an auxiliary adaptive weighted loss function (AWL) to automatically optimize the training process. MrsSeg first used a lightweight backbone to extract different-resolution features, then adopted a multi-resolution fusion module to fuse the local information and global information, and finally, a multi-level fusion decoder was used to aggregate and merge the features at different levels to get the desert segmentation result. In this method, each branch loss was treated as an independent task, AWL was proposed to calculate and adjust the weight of each branch. By giving priority to the easy tasks, the improved loss function could effectively improve the convergence speed of the model and the desert segmentation result. The experimental results showed that MrsSeg-AWL effectively improved the learning ability of the model and has faster convergence speed, lower parameter complexity, and more accurate segmentation results.