Automated three-dimensional image registration for longitudinal photoacoustic imaging

Significance

Photoacoustic tomography (PAT) has great potential in monitoring disease progression and treatment response in breast cancer. However, due to variations in breast repositioning, there is a chance of geometric misalignment between images. Further, poor repositioning can affect light fluence distribution and imaging field-of-view, making images different from one another. The net effect is that it becomes challenging to distinguish between image changes due to repositioning effects and those due to true biological variations.

Aim

The aim is to develop a three-dimensional image registration framework for geometrically aligning repeated PAT volumetric images, which are potentially affected by repositioning effects such as misalignment, changed radiant exposure conditions, and different fields-of-view.

Approach

The proposed framework involves the use of a coordinate-based neural network to represent the displacement field between pairs of PAT volumetric images. A loss function based on normalized cross correlation and Frangi vesselness feature extraction at multiple scales was implemented. We refer to our image registration framework as MUVINN-reg, which stands for multiscale vesselness-based image registration using neural networks. The approach was tested on a longitudinal dataset of healthy volunteer breast PAT images acquired with the hybrid photoacoustic-ultrasound Photoacoustic Mammoscope 3 imaging system. The registration performance was also tested under unfavorable repositioning conditions such as intentional mispositioning, and variation in breast-supporting cup size between measurements.

Results

A total of 13 pairs of repeated PAT scans were included in this study. MUVINN-reg showed excellent performance in co-registering each pair of images. The proposed framework was shown to be robust to image intensity shifts and field-of-view changes. Furthermore, MUVINN-reg could align vessels at imaging depths greater than 4 cm.

Conclusions

The proposed framework will enable the use of PAT for quantitative and reproducible monitoring of disease progression and treatment response.

Integration of Deep Learning and Active Shape Models for More Accurate Prostate Segmentation in 3D MR Images

Magnetic resonance imaging (MRI) has a growing role in the clinical workup of prostate cancer. However, manual three-dimensional (3D) segmentation of the prostate is a laborious and time-consuming task. In this scenario, the use of automated algorithms for prostate segmentation allows us to bypass the huge workload of physicians. In this work, we propose a fully automated hybrid approach for prostate gland segmentation in MR images using an initial segmentation of prostate volumes using a custom-made 3D deep network (VNet-T2), followed by refinement using an Active Shape Model (ASM). While the deep network focuses on three-dimensional spatial coherence of the shape, the ASM relies on local image information and this joint effort allows for improved segmentation of the organ contours. Our method is developed and tested on a dataset composed of T2-weighted (T2w) MRI prostatic volumes of 60 male patients. In the test set, the proposed method shows excellent segmentation performance, achieving a mean dice score and Hausdorff distance of 0.851 and 7.55 mm, respectively. In the future, this algorithm could serve as an enabling technology for the development of computer-aided systems for prostate cancer characterization in MR imaging.

From subjective to objective: A pilot study on testicular radiomics analysis as a measure of gonadal function

BACKGROUND

The connection between testicular ultrasound (US) parameters and testicular function, including both spermato- and steroidogenesis has been largely suggested, but their predictive properties are not routinely applied. Radiomics, a new engineering approach to radiological imaging, could overcome the visual limit of the sonographer.

OBJECTIVES

This study is aimed at extracting objective testicular US features, correlating with testicular function, including both spermato- and steroidogenesis, using an engineering approach, in order to overcome the operator-dependent subjectivity.

MATERIALS AND METHODS

Prospective observational pilot study from December 2019 to December 2020 on normozoospermic subjects and patients with semen variables alterations, excluding azoospermia. All patients underwent conventional semen analysis, pituitary-gonadal hormones assessment, and testicular US, performed by the same operator. US images were analyzed by Biolab (Turin) throughout image segmentation, image pre-processing, and texture features extraction.

RESULTS

One hundred seventy US testicular images were collected from 85 patients (age 38.6 ± 9.1 years). A total of 44 first-order and advanced features were extracted. US inhomogeneity defined by radiomics significantly correlates with the andrologist definition, showing for the first time a mathematical quantification of a subjective US evaluation. Thirteen US texture features correlated with semen parameters, predicting sperm concentration, total sperm number, progressive motility, total motility and morphology, and with gonadotropins serum levels, but not with total testosterone serum levels. Classification analyses confirmed that US textural features predicted patients' classification according to semen parameters alterations.

CONCLUSIONS

Radiomics texture features qualitatively describe the testicular parenchyma with objective and reliable quantitative parameters, reflecting both the testicular spermatogenic capability and the action of pituitary gonadotropins. This is an innovative model in which US texture features represent a mirror of the pituitary-gonadal homeostasis in terms of reproductive function.

Impact of segmentation and discretization on radiomic features in 68Ga-DOTA-TOC PET/CT images of neuroendocrine tumor

OBJECTIVE

To identify the impact of segmentation methods and intensity discretization on radiomic features (RFs) extraction from 68Ga-DOTA-TOC PET images in patients with neuroendocrine tumors.

METHODS

Forty-nine patients were retrospectively analyzed. Tumor contouring was performed manually by four different operators and with a semi-automatic edge-based segmentation (SAEB) algorithm. Three SUVmax fixed thresholds (20, 30, 40%) were applied. Fifty-one RFs were extracted applying two different intensity rescale factors for gray-level discretization: one absolute (AR60 = SUV from 0 to 60) and one relative (RR = min-max of the VOI SUV). Dice similarity coefficient (DSC) was calculated to quantify segmentation agreement between different segmentation methods. The impact of segmentation and discretization on RFs was assessed by intra-class correlation coefficients (ICC) and the coefficient of variance (COVL). The RFs' correlation with volume and SUVmax was analyzed by calculating Pearson's correlation coefficients.

RESULTS

DSC mean value was 0.75 ± 0.11 (0.45-0.92) between SAEB and operators and 0.78 ± 0.09 (0.36-0.97), among the four manual segmentations. The study showed high robustness (ICC > 0.9): (a) in 64.7% of RFs for segmentation methods using AR60, improved by applying SUVmax threshold of 40% (86.5%); (b) in 50.9% of RFs for different SUVmax thresholds using AR60; and (c) in 37% of RFs for discretization settings using different segmentation methods. Several RFs were not correlated with volume and SUVmax.

CONCLUSIONS

RFs robustness to manual segmentation resulted higher in NET 68Ga-DOTA-TOC images compared to 18F-FDG PET/CT images. Forty percent SUVmax thresholds yield superior RFs stability among operators, however leading to a possible loss of biological information. SAEB segmentation appears to be an optimal alternative to manual segmentation, but further validations are needed. Finally, discretization settings highly impacted on RFs robustness and should always be stated.

68Ga-DOTATOC PET/CT-Based Radiomic Analysis and PRRT Outcome: A Preliminary Evaluation Based on an Exploratory Radiomic Analysis on Two Patients

Aim: This work aims to evaluate whether the radiomic features extracted by 68Ga-DOTATOC-PET/CT of two patients are associated with the response to peptide receptor radionuclide therapy (PRRT) in patients affected by neuroendocrine tumor (NET).

Methods: This is a pilot report in two NET patients who experienced a discordant response to PRRT (responder vs. non-responder) according to RECIST1.1. The patients presented with liver metastasis from the rectum and pancreas G3-NET, respectively. Whole-body total-lesion somatostatin receptor-expression (TLSREwb-50) and somatostatin receptor-expressing tumor volume (SRETV wb-50) were obtained in pre- and post-PRRT PET/CT. Radiomic analysis was performed, extracting 38 radiomic features (RFs) from the patients' lesions. The Mann–Whitney test was used to compare RFs in the responder patient vs. the non-responder patient. Pearson correlation and principal component analysis (PCA) were used to evaluate the correlation and independence of the different RFs.

Results: TLSREwb-50 and SRETVwb-50 modifications correlate with RECIST1.1 response. A total of 28 RFs extracted on pre-therapy PET/CT showed significant differences between the two patients in the Mann–Whitney test (p < 0.05). A total of seven second-order features, with poor correlation with SUVmax and PET volume, were identified by the Pearson correlation matrix. Finally, the first two PCA principal components explain 83.8% of total variance.

Conclusion: TLSREwb-50 and SRETVwb-50 are parameters that might be used to predict and to assess the PET response to PRRT. RFs might have a role in defining inter-patient heterogeneity and in the prediction of therapy response. It is important to implement future studies with larger and more homogeneous patient populations to confirm the efficacy of these biomarkers.

Automatic segmentation of ultrasound images of gastrocnemius medialis with different echogenicity levels using convolutional neural networks

The purpose of this study was to develop an automatic method for the segmentation of muscle cross-sectional area on transverse B-mode ultrasound images of gastrocnemius medialis using a convolutional neural network(CNN). In the provided dataset images with both normal and increased echogenicity are present. The manually annotated dataset consisted of 591 images, from 200 subjects, 400 relative to subjects with normal echogenicity and 191 to subjects with augmented echogenicity. From the DICOM files, the image has been extracted and processed using the CNN, then the output has been post-processed to obtain a finer segmentation. Final results have been compared to the manual segmentations. Precision and Recall scores as mean ± standard deviation for training, validation, and test sets are 0.96 ± 0.05, 0.90 ± 0.18, 0.89 ± 0.15 and 0.97 ±0.03, 0.89± 0.17, 0.90 ± 0.14 respectively. The CNN approach has also been compared to another automatic algorithm, showing better performances. The proposed automatic method provides an accurate estimation of muscle cross-sectional area in muscles with different echogenicity levels.

Automatic Extraction of Dermatological Parameters from Nevi Using an Inexpensive Smartphone Microscope: A Proof of Concept

The evolution of smartphone technology has made their use more common in dermatological applications. Here we studied the feasibility of using an inexpensive smartphone microscope for the extraction of dermatological parameters and compared the results obtained with a portable dermoscope, commonly used in clinical practice. Forty-two skin lesions were imaged with both devices and visually analyzed by an expert dermatologist. The presence of a reticular pattern was observed in 22 dermoscopic images, but only in 10 smartphone images. The proposed paradigm segments the image and extracts texture features which are used to train and validate a neural network to classify the presence of a reticular pattern. Using 5-fold cross-validation, an accuracy of 100% and 95% was obtained with the dermoscopic and smartphone images, respectively. This approach can be useful for general practitioners and as a triage tool for skin lesion analysis.

Non-invasive analysis of actinic keratosis using a cold stimulation and near-infrared spectroscopy

Non-melanoma skin cancers are the most common tumor in the Caucasian population, and include actinic keratosis (AK), which is considered an early form of in-situ squamous cell carcinoma (SCC). Currently the only way to monitor lesion progression (i.e., from AK to invasive SCC) is through an invasive bioptic procedure. Near-infrared spectroscopy (NIRS) is a non-invasive technique that studies haemoglobin (oxygenated haemoglobin, O2Hb, and deoxygenated haemoglobin, HHb) relative concentration variations. The objective of this study is to evaluate if AKs present a different vascular response when compared to healthy skin using time and frequency parameters extracted from the NIRS signals. The NIRS signals were acquired on the AKs and a healthy skin area of patients (n=53), with the same acquisition protocol: baseline signals (1.5 min), application of ice pack near lesion (1.5 min), removal of ice pack and acquisition of vascular recovery (1.5 min). We calculated 18 features to evaluate if the vascular response was different in the two cases (i.e., healthy skin and AK lesions). By applying the multivariate analysis of variance (MANOVA), a statistically significant difference is found in the O2Hb and HHb after the stimulus application. This shows how the NIRS technique can give important vascular information that could help the diagnosis of a lesion and the evaluation of its progression. Overall, the obtained results encourage us to look further into the study of the skin lesions and their progression with NIRS signals.

Multimodal T2w and DWI Prostate Gland Automated Registration

Multiparametric magnetic resonance imaging (mpMRI) is emerging as a promising tool in the clinical pathway of prostate cancer (PCa). The registration between a structural and a functional imaging modality, such as T2-weighted (T2w) and diffusion-weighted imaging (DWI) is fundamental in the development of a mpMRI-based computer aided diagnosis (CAD) system for PCa. Here, we propose an automated method to register the prostate gland in T2w and DWI image sequences by a landmark-based affine registration and a non-parametric diffeomorphic registration. An expert operator manually segmented the prostate gland in both modalities on a dataset of 20 patients. Target registration error and Jaccard index, which measures the overlap between masks, were evaluated pre- and post- registration resulting in an improvement of 44% and 21%, respectively. In the future, the proposed method could be useful in the framework of a CAD system for PCa detection and characterization in mpMRI.

Transverse Muscle Ultrasound Analysis (TRAMA): Robust and Accurate Segmentation of Muscle Cross-Sectional Area

Ultrasonography allows non-invasive and real time-measurement of the visible cross-sectional area (CSA) of muscles, which is a clinically relevant descriptor of muscle size. The aim of this study was to develop and validate a fully automatic method called transverse muscle ultrasound analysis (TRAMA) for segmentation of the muscle in B-mode transverse ultrasound images and measurement of muscle CSA. TRAMA was tested on a database of 200 ultrasound images of the rectus femoris, vastus lateralis, tibialis anterior and medial gastrocnemius muscles. The automatic CSA measurements were compared with manual measurements obtained by two operators. There were no statistical differences between the automatic and manual measurements of CSA of the four muscles, and TRAMA performance was comparable to intra-operator variability in terms of the Dice similarity coefficient and Hausdorff distance between the automatic and manual segmentations. Compared with manual segmentation, the Dice similarity coefficient for the proposed method was always higher than 93%; the Hausdorff distance never exceeded 4 mm, and the maximum absolute error was 62 mm². TRAMA is the first automated algorithm that analyzes and segments ultrasound scans of the muscle in the transverse plane. It can be adopted in future studies for automatic segmentation of muscle regions of interest to enhance and automatize a multitexture analysis of muscle structure.