Joint Detection of Tap and CEA Based on Deep Learning Medical Image Segmentation: Risk Prediction of Thyroid Cancer

In recent years, the incidence of thyroid nodules has shown an increasing trend year by year and has become one of the important diseases that endanger human health. Ultrasound medical images based on deep learning are widely used in clinical diagnosis due to their cheapness, no radiation, and low cost. The use of image processing technology to accurately segment the nodule area provides important auxiliary information for the doctor’s diagnosis, which is of great value for guiding clinical treatment. The purpose of this article is to explore the application value of combined detection of abnormal sugar-chain glycoprotein (TAP) and carcinoembryonic antigen (CEA) in the risk estimation of thyroid cancer in patients with thyroid nodules of type IV and above based on deep learning medical images. In this paper, ultrasound thyroid images are used as the research content, and the active contour level set method is used as the segmentation basis, and a segmentation algorithm for thyroid nodules is proposed. This paper takes ultrasound thyroid images as the research content, uses the active contour level set method as the basis of segmentation, and proposes an image segmentation algorithm Fast-SegNet based on deep learning, which extends the network model that was mainly used for thyroid medical image segmentation to more scenarios of the segmentation task. From January 2019 to October 2020, 400 patients with thyroid nodules of type IV and above were selected for physical examination and screening at the Health Management Center of our hospital, and they were diagnosed as thyroid cancer by pathological examination of thyroid nodules under B-ultrasound positioning. The detection rates of thyroid cancer in patients with thyroid nodules of type IV and above are compared; serum TAP and CEA levels are detected; PT-PCR is used to detect TTF-1, PTEN, and NIS expression; the detection, missed diagnosis, misdiagnosis rate, and diagnostic efficiency of the three detection methods are compared. This article uses the thyroid nodule region segmented based on deep learning medical images and compares experiments with CV model, LBF model, and DRLSE model. The experimental results show that the segmentation overlap rate of this method is as high as 98.4%, indicating that the algorithm proposed in this paper can more accurately extract the thyroid nodule area.

Perceptual Asymmetries and Auditory Processing of Estonian Quantities

Similar to visual perception, auditory perception also has a clearly described “pop-out” effect, where an element with some extra feature is easier to detect among elements without an extra feature. This phenomenon is better known as auditory perceptual asymmetry. We investigated such asymmetry between shorter or longer duration, and level or falling of pitch of linguistic stimuli that carry a meaning in one language (Estonian), but not in another (Russian). For the mismatch negativity (MMN) experiment, we created four different types of stimuli by modifying the duration of the first vowel [ɑ] (170, 290 ms) and pitch contour (level vs. falling pitch) of the stimuli words (‘SATA,’ ‘SAKI’). The stimuli were synthesized from Estonian words (‘SATA,’ ‘SAKI’) and follow the Estonian language three-way quantity system, which incorporates tonal features (falling pitch contour) together with temporal patterns. This made the meaning of the word dependent on the combination of both features and allows us to compare the relative contribution of duration and pitch contour in discrimination of language stimuli in the brain via MMN generation. The participants of the experiment were 12 Russian native speakers with little or no experience in Estonian and living in Estonia short-term, and 12 Estonian native speakers (age 18–27 years). We found that participants’ perception of the linguistic stimuli differed not only according to the physical features but also according to their native language, confirming that the meaning of the word interferes with the early automatic processing of phonological features. The GAMM and ANOVA analysis of the reversed design results showed that the deviant with longer duration among shorter standards elicited a MMN response with greater amplitude than the short deviant among long standards, while changes in pitch contour (falling vs. level pitch) produced neither strong MMN nor asymmetry. Thus, we demonstrate the effect of language background on asymmetric perception of linguistic stimuli that aligns with those of previous studies (Jaramillo et al., 2000), and contributes to the growing body of knowledge supporting auditory perceptual asymmetry.

Contour Gradient Optimization

Inspired by local cooperation in the real world, a new evolutionary algorithm, Contour Gradient Optimization algorithm (CGO), is proposed for solving optimization problems. CGO is a new type of population-based algorithm that emulates the cooperation among neighbors. Each individual in CGO evolves in its neighborhood environment to find a better region. Each individual moves with a velocity measured by the field of its nearest individuals. The field includes the attractive forces from its better neighbor in the higher contour level and the repulsive force from its worse neighbor in the lower contour level. In this chapter, CGO is compared with six different widely used optimization algorithms, and comparative analysis shows that CGO is better than these algorithms in respect of accuracy and effectiveness.

Strategic Noise Mapping Prediction for a Rubber Manufacturing Factory in Malaysia

Rubber product manufacturing industry was found with severe occupational noise exposure problems due to improper noise management and lack of reliable noise information at the workplace. Strategic noise mapping provides important information in monitoring the occupational noise. Therefore, a case study was conducted to investigate the current noise exposure circumstances based on the information from the noise map and noise risk zones. The stochastic noise mapping simulation method was applied to predict these maps. Based on the results, most of the regions in the operation area were bounded with a noise contour level of 80 dBA and some small regions were exceeded the noise level of 100 dBA. More than 45 % of mapping area was categorised as extremely high risk and high risk zones. Workers are exposed to the high noise level in this workplace. The management should take immediate action for controlling noise and always supervise their workers in using the hearing protection equipment.

Feature-Based Uncertainty Visualization

While uncertainty in scientific data attracts an increasing research interest in the visualization community, two critical issues remain insufficiently studied: (1) visualizing the impact of the uncertainty of a data set on its features and (2) interactively exploring 3D or large 2D data sets with uncertainties. In this chapter, a suite of feature-based techniques is developed to address these issues. First, an interactive visualization tool for exploring scalar data with data-level, contour-level, and topology-level uncertainties is developed. Second, a framework of visualizing feature-level uncertainty is proposed to study the uncertain feature deviations in both scalar and vector data sets. With quantified representation and interactive capability, the proposed feature-based visualizations provide new insights into the uncertainties of both data and their features which otherwise would remain unknown with the visualization of only data uncertainties.

Feature-Based Uncertainty Visualization

While uncertainty in scientific data attracts an increasing research interest in the visualization community, two critical issues remain insufficiently studied: (1) visualizing the impact of the uncertainty of a data set on its features and (2) interactively exploring 3D or large 2D data sets with uncertainties. In this chapter, a suite of feature-based techniques is developed to address these issues. First, an interactive visualization tool for exploring scalar data with data-level, contour-level, and topology-level uncertainties is developed. Second, a framework of visualizing feature-level uncertainty is proposed to study the uncertain feature deviations in both scalar and vector data sets. With quantified representation and interactive capability, the proposed feature-based visualizations provide new insights into the uncertainties of both data and their features which otherwise would remain unknown with the visualization of only data uncertainties.

Contour Gradient Optimization

Inspired by the local cooperation behavior in the real world, a new evolutionary algorithm Contour Gradient Optimization algorithm (CGO) is proposed for solving optimization problems. CGO is a new type of global search algorithm that emulates the cooperation among neighbors. Each individual in CGO evolves in its neighborhood environment to find a better region. Each individual moves with a velocity measured by the field of its nearest individuals. The field includes the attractive forces from its better neighbor in the higher contour level and the repulsive force from its worse neighbor in the lower contour level. Intensive simulations were performed and the results show that CGO is able to solve the tested multimodal optimization problems globally. In this paper, CGO is thoroughly compared with six different widely used optimization algorithms under sixteen different benchmark functions. The comparative analysis shows that CGO is comparatively better than these algorithms in the respect of accuracy and effectiveness.