Photoacoustic imaging (PAI) has the potential to detect cancer in the early stage. PAI is safe due to its non-ionizing radiation properties, which greatly enhance its clinical feasibility in the near future, which provides significant benefits over other imaging techniques like X-ray computed tomography (CT). In this paper, the fully automated 3D deep learning cancer detector is taken to detect and localize the presence of cancer in freshly excised ex-vivo human thyroid and prostate tissue specimens using a three-dimensional (3D) multispectral photoacoustic (MPA) dataset automatically. The model detected and localized the cancer region in a given test MPA image with promising results.
Multispectral photoacoustic (MPA) specimen imaging modality is proven successful in differentiating photoacoustic (PA) signal characteristics from a cancer and normal region. The oxy and de-oxy hemoglobin content in a human tissue captured in the MPA data are the key features for cancer detection. In this study, we propose to use deep 3D convolution neural network trained on the thyroid MPA dataset and tested on the prostate MPA dataset to evaluate this potential. The proposed algorithm first extracts the spatial, spectral, and temporal features from the thyroid MPA image data using 3D convolutional layers and detects cancer tissue using the logistic function, the last layer of the network. The model achieved an AUC (area under the curve) of the ROC (receiver operating characteristic) curve of 0.72 on the prostate MPA dataset.
Pathology diagnosis is usually done by a human pathologist observing tissue stained glass slide under a microscope. In the case of multi-specimen study to locate cancer region, such as in thyroidectomy, significant labor-intensive processing is required at high cost. Multispectral photoacoustic (MPA) specimen imaging, has proven successful in differentiating photoacoustic (PA) signal characteristics between a histopathology defined cancer region and normal tissue. A more pragmatic research question to ask is, can MPA imaging data predict, whether a sectioned tissue slice has cancer region(s)? We propose to use inception-resnet-v2 convolutional neural networks (CNNs) on the thyroid MPA data to evaluate this potential by transfer learning. The proposed algorithm first extracts features from the thyroid MPA image data using CNN and then detects cancer using the softmax function, the last layer of the network. The AUCs (area under curve) of the receiver operating characteristic (ROC) curve of cancer, benign nodule and normal are 0.73, 0.81, and 0.88 respectively with a limited number of the MPA dataset.
This paper evaluates the detection performance of the three subpixel target detection algorithms based on the spectral signature of a target. Three subpixel target detection algorithms, Adaptive Coherence Estimator (ACE), Spectral Matched Filter (SMF), and Constrained Energy Minimization (CEM) are evaluated and compared using the Principal Component Analysis (PCA) spaced RIT Avon12 hyperspectral dataset. The performance of the three detectors is evaluated by generating the Receiver Operating Characteristic (ROC) curve. The ROC curves are generated by uploading the detection statistics image produced by the three detectors to the Data and Algorithm Standard Evaluation ( DASE) Website of IEEE Geoscience and Remote Sensing Society(GRSS) . Finally, we note the Area Under Curve (AUC) as the proposed utility metric value to evaluate the performance of the three detectors. The AUCs of the ROC curve produced by the ACE, CEM, and SMF are 94.0 %, 93.9 %, and 87.2 % respectively.
Intracranial hemorrhage is a critical conditional with the high mortality rate that is typically diagnosed based on head computer tomography (CT) images. Deep learning algorithms, in particular, convolution neural networks (CNN), are becoming the methodology of choice in medical image analysis for a variety of applications such as computer-aided diagnosis, and segmentation. In this study, we propose a fully automated deep learning framework which learns to detect brain hemorrhage based on cross sectional CT images. The dataset for this work consists of 40,367 3D head CT studies (over 1.5 million 2D images) acquired retrospectively over a decade from multiple radiology facilities at Geisinger Health System. The proposed algorithm first extracts features using 3D CNN and then detects brain hemorrhage using the logistic function as the last layer of the network. Finally, we created an ensemble of three different 3D CNN architectures to improve the classification accuracy. The area under the curve (AUC) of the receiver operator characteristic (ROC) curve of the ensemble of three architectures was 0.87. Their results are very promising considering the fact that the head CT studies were not controlled for slice thickness, scanner type, study protocol or any other settings. Moreover, the proposed algorithm reliably detected various types of hemorrhage within the skull. This work is one of the first applications of 3D CNN trained on a large dataset of cross sectional medical images for detection of a critical radiological condition
KEYWORDS: Continuous wave operation, Pulsed laser operation, Signal generators, Photoacoustic spectroscopy, MATLAB, Signal to noise ratio, Ultrasonography, Real time imaging, Optical simulations, Semiconductor lasers
Photoacoustic (PA) imaging is a hybrid imaging modality that integrates the strength of optical and ultrasound imaging. Nanosecond (ns) pulsed lasers used in current PA imaging systems are expensive, bulky and they often waste energy. We propose and evaluate, through simulations, the use of a continuous wave (CW) laser whose amplitude is linear frequency modulated (chirp) for PA imaging. The chirp signal provides signal-to-side-lobe ratio (SSR) improvement potential and full control over PA signal frequencies excited in the sample. The PA signal spectrum is a function of absorber size and the time frequencies present in the chirp. A mismatch between the input chirp spectrum and the output PA signal spectrum can affect the compressed pulse that is recovered from cross-correlating the two. We have quantitatively characterized this effect. The k-wave Matlab tool box was used to simulate PA signals in three dimensions for absorbers ranging in size from 0.1 mm to 0.6 mm, in response to laser excitation amplitude that is linearly swept from 0.5 MHz to 4 MHz. This sweep frequency range was chosen based on the spectrum analysis of a PA signal generated from ex-vivo human prostate tissue samples. In comparison, the energy wastage by a ns laser pulse was also estimated. For the chirp methodology, the compressed pulse peak amplitude, pulse width and side lobe structure parameters were extracted for different size absorbers. While the SSR increased 6 fold with absorber size, the pulse width decreased by 25%.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.