Fine-grained classification of cervical cells into different abnormality levels is of great clinical importance but remains very challenging. Contrary to traditional classification methods that rely on handcrafted or engineered features, convolution neural network (CNN) can classify cervical cells based on automatically learned deep features. However, CNN in previous studies do not involve cell morphological information, and it is unknown whether morphological features can be directly modeled by CNN to classify cervical cells. This paper presents a CNN-based method that combines cell image appearance with cell morphology for classification of cervical cells in Pap smear. The training cervical cell dataset consists of adaptively re-sampled image patches coarsely centered on the nuclei. Several CNN models (AlexNet, GoogleNet, ResNet and DenseNet) pre-trained on ImageNet dataset were fine-tuned on the cervical dataset for comparison. The proposed method is evaluated on the Herlev cervical dataset by five-fold cross-validation at patient level splitting. Results show that by adding cytoplasm and nucleus masks as raw morphological information into appearance-based CNN learning, higher classification accuracies can be achieved in general. Among the four CNN models, GoogleNet fed with both morphological and appearance information obtains the highest classification accuracies of 94.5% for 2-class classification task and 64.5% for 7-class classification task. Our method demonstrates that combining cervical cell morphology with appearance information can provide improved classification performance, which is clinically important for early diagnosis of cervical dysplastic changes.
As magneto-acoustic-electrical tomography (MAET) combines the merits of high contrast and high imaging resolution, and is extremely useful for electrical conductivity measurement, so it is expected to be a promising medical imaging modalities for diagnosis of early-stage cancer. Based on the Verasonics system and the MC600 displacement platform, we designed and implemented a MAET system with a chirp pulse stimulation (MAET-CPS) method and a focal probe was utilized for stepscan focus excitation to enhance the imaging resolution. The relevant experiments were conducted to explore the influence of excitation positions of the single-focus point, and the effect of the excitation position on the amplitudes of the conductivity variation was clearly demonstrated. In order to take advantage of the merits of multifocus imaging, we firstly proposed a single focus MAET system with a chirp pulse stimulation (sfMAET-CPS) method and a multifocus MAET system with a chirp pulse stimulation (mfMAET-CPS) method for high-resolution conductivity imaging, and a homogenous gelatin phantom with a cuboid-shaped hole was used to investigate the accuracy of mfMAET-CPS. Comparative experiments were carried out on the same uniform phantom by the sfMAET-CPS and the mfMAET-CPS, respectively. The results showed that: (1) the electrical conductivity distributions of the homogenous phantom with a cuboid-shaped hole were detected by the sfMAET-CPS but were easily affected by the focal point, which demonstrated that the sfMAET-CPS had a low imaging resolution. (2) Compared with the sfMAET-CPS, the imaging effect of the mfMAET-CPS was much better than that of the sfMAET-CPS. (3) A linear interpolation algorithm was used to process the 2D conductivity distribution; it increased the smoothness of the conductivity distribution and improved the imaging effect. The stepscan focus excitation and the linearly frequency-modulated theory provide an alternative scheme for the clinical application of MAET.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.