Target depth-regularized reconstruction in diffuse optical tomography using ultrasound segmentation as prior information

Automated pipeline for breast cancer diagnosis using US assisted diffuse optical tomography

Ultrasound (US)-guided diffuse optical tomography (DOT) is a portable and non-invasive imaging modality for breast cancer diagnosis and treatment response monitoring. However, DOT data pre-processing and imaging reconstruction often require labor intensive manual processing which hampers real-time diagnosis. In this study, we aim at providing an automated US-assisted DOT pre-processing, imaging and diagnosis pipeline to achieve near real-time diagnosis. We have developed an automated DOT pre-processing method including motion detection, mismatch classification using deep-learning approach, and outlier removal. US-lesion information needed for DOT reconstruction was extracted by a semi-automated lesion segmentation approach combined with a US reading algorithm. A deep learning model was used to evaluate the quality of the reconstructed DOT images and a two-step deep-learning model developed earlier is implemented to provide final diagnosis based on US imaging features and DOT measurements and imaging results. The presented US-assisted DOT pipeline accurately processed the DOT measurements and reconstruction and reduced the procedure time to 2 to 3 minutes while maintained a comparable classification result with manually processed dataset.

Two-stage classification strategy for breast cancer diagnosis using ultrasound-guided diffuse optical tomography and deep learning

Significance

Ultrasound (US)-guided diffuse optical tomography (DOT) has demonstrated great potential for breast cancer diagnosis in which real-time or near real-time diagnosis with high accuracy is desired.

Aim

We aim to use US-guided DOT to achieve an automated, fast, and accurate classification of breast lesions.

Approach

We propose a two-stage classification strategy with deep learning. In the first stage, US images and histograms created from DOT perturbation measurements are combined to predict benign lesions. Then the non-benign suspicious lesions are passed through to the second stage, which combine US image features, DOT histogram features, and 3D DOT reconstructed images for final diagnosis.

Results

The first stage alone identified 73.0% of benign cases without image reconstruction. In distinguishing between benign and malignant breast lesions in patient data, the two-stage classification approach achieved an area under the receiver operating characteristic curve of 0.946, outperforming the diagnoses of all single-modality models and of a single-stage classification model that combines all US images, DOT histogram, and imaging features.

Conclusions

The proposed two-stage classification strategy achieves better classification accuracy than single-modality-only models and a single-stage classification model that combines all features. It can potentially distinguish breast cancers from benign lesions in near real-time.

Resolution and penetration depth of reflection-mode time-domain near infrared optical tomography using a ToF SPAD camera

In a turbid medium such as biological tissue, near-infrared optical tomography (NIROT) can image the oxygenation, a highly relevant clinical parameter. To be an efficient diagnostic tool, NIROT has to have high spatial resolution and depth sensitivity, fast acquisition time, and be easy to use. Since many tissues cannot be penetrated by near-infrared light, such tissue needs to be measured in reflection mode, i.e., where light emission and detection components are placed on the same side. Thanks to the recent advance in single-photon avalanche diode (SPAD) array technology, we have developed a compact reflection-mode time-domain (TD) NIROT system with a large number of channels, which is expected to substantially increase the resolution and depth sensitivity of the oxygenation images. The aim was to test this experimentally for our SPAD camera-empowered TD NIROT system. Experiments with one and two inclusions, i.e., optically dense spheres of 5mm radius, immersed in turbid liquid were conducted. The inclusions were placed at depths from 10mm to 30mm and moved across the field-of-view. In the two-inclusion experiment, two identical spheres were placed at a lateral distance of 8mm. We also compared short exposure times of 1s, suitable for dynamic processes, with a long exposure of 100s. Additionally, we imaged complex geometries inside the turbid medium, which represented structural elements of a biological object. The quality of the reconstructed images was quantified by the root mean squared error (RMSE), peak signal-to-noise ratio (PSNR), and dice similarity. The two small spheres were successfully resolved up to a depth of 30mm. We demonstrated robust image reconstruction even at 1s exposure. Furthermore, the complex geometries were also successfully reconstructed. The results demonstrated a groundbreaking level of enhanced performance of the NIROT system based on a SPAD camera.

Difference imaging from single measurements in diffuse optical tomography: a deep learning approach

SIGNIFICANCE

"Difference imaging," which reconstructs target optical properties using measurements with and without target information, is often used in diffuse optical tomography (DOT) in vivo imaging. However, taking additional reference measurements is time consuming, and mismatches between the target medium and the reference medium can cause inaccurate reconstruction.

AIM

We aim to streamline the data acquisition and mitigate the mismatch problems in DOT difference imaging using a deep learning-based approach to generate data from target measurements only.

APPROACH

We train an artificial neural network to output data for difference imaging from target measurements only. The model is trained and validated on simulation data and tested with simulations, phantom experiments, and clinical data from 56 patients with breast lesions.

RESULTS

The proposed method has comparable performance to the traditional approach using measurements without mismatch between the target side and the reference side, and it outperforms the traditional approach using measurements when there is a mismatch. It also improves the target-to-artifact ratio and lesion localization in patient data.

CONCLUSIONS

The proposed method can simplify the data acquisition procedure, mitigate mismatch problems, and improve reconstructed image quality in DOT difference imaging.

Machine learning model with physical constraints for diffuse optical tomography

A machine learning model with physical constraints (ML-PC) is introduced to perform diffuse optical tomography (DOT) reconstruction. DOT reconstruction is an ill-posed and under-determined problem, and its quality suffers by model mismatches, complex boundary conditions, tissue-probe contact, noise etc. Here, for the first time, we combine ultrasound-guided DOT with ML to facilitate DOT reconstruction. Our method has two key components: (i) a neural network based on auto-encoder is adopted for DOT reconstruction, and (ii) physical constraints are implemented to achieve accurate reconstruction. Both qualitative and quantitative results demonstrate that the accuracy of the proposed method surpasses that of existing models. In a phantom study, compared with the Born conjugate gradient descent (Born-CGD) reconstruction method, the ML-PC method decreases the mean percentage error of the reconstructed maximum absorption coefficient from 16.41% to 13.4% for high contrast phantoms and from 23.42% to 9.06% for low contrast phantoms, with improved depth distribution of the target absorption maps. In a clinical study, better contrast was obtained between malignant and benign breast lesions, with the ratio of the medians of the maximum absorption coefficient improved from 1.63 to 2.22.

Ultrasound segmentation-guided edge artifact reduction in diffuse optical tomography using connected component analysis

Ultrasound (US)-guided diffuse optical tomography (DOT) has demonstrated potential value for breast cancer diagnosis and treatment response assessment. However, in clinical use, the chest wall, poor probe-tissue contact, and tissue heterogeneity can all cause image artifacts. These image artifacts, appearing commonly as hot spots in the non-lesion regions (edge artifacts), can decrease the reconstruction accuracy and cause misinterpretation of lesion images. Here we introduce an iterative, connected component analysis-based image artifact reduction algorithm. A convolutional neural network (CNN) is used to segment co-registered US images to extract the lesion location and size to guide the artifact reduction. We demonstrate its performance using Monte Carlo simulations on VICTRE digital breast phantoms and breast patient images. In simulated tissue mismatch models, this algorithm successfully reduces edge artifacts without significantly changing the reconstructed target absorption coefficients. With clinical data it improves the optical contrast between malignant and benign groups, from 1.55 without artifact reduction to 1.91 with artifact reduction. The proposed algorithm has a broad range of applications in other modality-guided DOT imaging.