Microscope-integrated optical coherence tomography for in vivo human brain tumor detection with artificial intelligence

OBJECTIVE

It has been shown that optical coherence tomography (OCT) can identify brain tumor tissue and potentially be used for intraoperative margin diagnostics. However, there is limited evidence on its use in human in vivo settings, particularly in terms of its applicability and accuracy of residual brain tumor detection (RTD). For this reason, a microscope-integrated OCT system was examined to determine in vivo feasibility of RTD after resection with automated scan analysis.

METHODS

Healthy and diseased brain was 3D scanned at the resection edge in 18 brain tumor patients and investigated for its informative value in regard to intraoperative tissue classification. Biopsies were taken at these locations and labeled by a neuropathologist for further analysis as ground truth. Optical OCT properties were obtained, compared, and used for separation with machine learning. In addition, two artificial intelligence-assisted methods were utilized for scan classification, and all approaches were examined for RTD accuracy and compared to standard techniques.

RESULTS

In vivo OCT tissue scanning was feasible and easily integrable into the surgical workflow. Measured backscattered light signal intensity, signal attenuation, and signal homogeneity were significantly distinctive in the comparison of scanned white matter to increasing levels of scanned tumor infiltration (p < 0.001) and achieved high values of accuracy (85%) for the detection of diseased brain in the tumor margin with support vector machine separation. A neuronal network approach achieved 82% accuracy and an autoencoder approach 85% accuracy in the detection of diseased brain in the tumor margin. Differentiating cortical gray matter from tumor tissue was not technically feasible in vivo.

CONCLUSIONS

In vivo OCT scanning of the human brain has been shown to contain significant value for intraoperative RTD, supporting what has previously been discussed for ex vivo OCT brain tumor scanning, with the perspective of complementing current intraoperative methods for this purpose, especially when deciding to withdraw from further resection toward the end of the surgery.

Application of deep learning to pressure injury staging

OBJECTIVE

Accurate assessment of pressure injuries (PIs) is necessary for a good outcome. Junior and non-specialist nurses have less experience with PIs and lack clinical practice, and so have difficulty staging them accurately. In this work, a deep learning-based system for PI staging and tissue classification is proposed to help improve its accuracy and efficiency in clinical practice, and save healthcare costs.

METHOD

A total of 1610 cases of PI and their corresponding photographs were collected from clinical practice, and each sample was accurately staged and the tissues labelled by experts for training a Mask Region-based Convolutional Neural Network (Mask R-CNN, Facebook Artificial Intelligence Research, Meta, US) object detection and instance segmentation network. A recognition system was set up to automatically stage and classify the tissues of the remotely uploaded PI photographs.

RESULTS

On a test set of 100 samples, the average precision of this model for stage recognition reached 0.603, which exceeded that of the medical personnel involved in the comparative evaluation, including an enterostomal therapist.

CONCLUSION

In this study, the deep learning-based PI staging system achieved the evaluation performance of a nurse with professional training in wound care. This low-cost system could help overcome the difficulty of identifying PIs by junior and non-specialist nurses, and provide valuable auxiliary clinical information.

Margin assessment during breast conserving surgery using diffuse reflectance spectroscopy

Significance

During breast-conserving surgeries, it is essential to evaluate the resection margins (edges of breast specimen) to determine whether the tumor has been removed completely. In current surgical practice, there are no methods available to aid in accurate real-time margin evaluation.

Aim

In this study, we investigated the diagnostic accuracy of diffuse reflectance spectroscopy (DRS) combined with tissue classification models in discriminating tumorous tissue from healthy tissue up to 2 mm in depth on the actual resection margin of in vivo breast tissue.

Approach

We collected an extensive dataset of DRS measurements on ex vivo breast tissue and in vivo breast tissue, which we used to develop different classification models for tissue classification. Next, these models were used in vivo to evaluate the performance of DRS for tissue discrimination during breast conserving surgery. We investigated which training strategy yielded optimum results for the classification model with the highest performance.

Results

We achieved a Matthews correlation coefficient of 0.76, a sensitivity of 96.7% (95% CI 95.6% to 98.2%), a specificity of 90.6% (95% CI 86.3% to 97.9%) and an area under the curve of 0.98 by training the optimum model on a combination of ex vivo and in vivo DRS data.

Conclusions

DRS allows real-time margin assessment with a high sensitivity and specificity during breast-conserving surgeries.

Review of machine learning for optical imaging of burn wound severity assessment

Significance

Over the past decade, machine learning (ML) algorithms have rapidly become much more widespread for numerous biomedical applications, including the diagnosis and categorization of disease and injury.

Aim

Here, we seek to characterize the recent growth of ML techniques that use imaging data to classify burn wound severity and report on the accuracies of different approaches.

Approach

To this end, we present a comprehensive literature review of preclinical and clinical studies using ML techniques to classify the severity of burn wounds.

Results

The majority of these reports used digital color photographs as input data to the classification algorithms, but recently there has been an increasing prevalence of the use of ML approaches using input data from more advanced optical imaging modalities (e.g., multispectral and hyperspectral imaging, optical coherence tomography), in addition to multimodal techniques. The classification accuracy of the different methods is reported; it typically ranges from ∼ 70 % to 90% relative to the current gold standard of clinical judgment.

Conclusions

The field would benefit from systematic analysis of the effects of different input data modalities, training/testing sets, and ML classifiers on the reported accuracy. Despite this current limitation, ML-based algorithms show significant promise for assisting in objectively classifying burn wound severity.

Toward the use of diffuse reflection spectroscopy for intra-operative tissue discrimination during sarcoma surgery

Significance

Accurately distinguishing tumor tissue from normal tissue is crucial to achieve complete resections during soft tissue sarcoma (STS) surgery while preserving critical structures. Incomplete tumor resections are associated with an increased risk of local recurrence and worse patient prognosis.

Aim

We evaluate the performance of diffuse reflectance spectroscopy (DRS) to distinguish tumor tissue from healthy tissue in STSs.

Approach

DRS spectra were acquired from different tissue types on multiple locations in 20 freshly excised sarcoma specimens. A k -nearest neighbors classification model was trained to predict the tissue types of the measured locations, using binary and multiclass approaches.

Results

Tumor tissue could be distinguished from healthy tissue with a classification accuracy of 0.90, sensitivity of 0.88, and specificity of 0.93 when well-differentiated liposarcomas were included. Excluding this subtype, the classification performance increased to an accuracy of 0.93, sensitivity of 0.94, and specificity of 0.93. The developed model showed a consistent performance over different histological subtypes and tumor locations.

Conclusions

Automatic tissue discrimination using DRS enables real-time intra-operative guidance, contributing to more accurate STS resections.

Clustering Methods for Vibro-Acoustic Sensing Features as a Potential Approach to Tissue Characterisation in Robot-Assisted Interventions

This article provides a comprehensive analysis of the feature extraction methods applied to vibro-acoustic signals (VA signals) in the context of robot-assisted interventions. The primary objective is to extract valuable information from these signals to understand tissue behaviour better and build upon prior research. This study is divided into three key stages: feature extraction using the Cepstrum Transform (CT), Mel-Frequency Cepstral Coefficients (MFCCs), and Fast Chirplet Transform (FCT); dimensionality reduction employing techniques such as Principal Component Analysis (PCA), t-Distributed Stochastic Neighbour Embedding (t-SNE), and Uniform Manifold Approximation and Projection (UMAP); and, finally, classification using a nearest neighbours classifier. The results demonstrate that using feature extraction techniques, especially the combination of CT and MFCC with dimensionality reduction algorithms, yields highly efficient outcomes. The classification metrics (Accuracy, Recall, and F1-score) approach 99%, and the clustering metric is 0.61. The performance of the CT-UMAP combination stands out in the evaluation metrics.

Shedding Light on Colorectal Cancer: An In Vivo Raman Spectroscopy Approach Combined with Deep Learning Analysis

Raman spectroscopy has emerged as a powerful tool in medical, biochemical, and biological research with high specificity, sensitivity, and spatial and temporal resolution. Recent advanced Raman systems, such as portable Raman systems and fiber-optic probes, provide the potential for accurate in vivo discrimination between healthy and cancerous tissues. In our study, a portable Raman probe spectrometer was tested in immunosuppressed mice for the in vivo localization of colorectal cancer malignancies from normal tissue margins. The acquired Raman spectra were preprocessed, and principal component analysis (PCA) was performed to facilitate discrimination between malignant and normal tissues and to highlight their biochemical differences using loading plots. A transfer learning model based on a one-dimensional convolutional neural network (1D-CNN) was employed for the Raman spectra data to assess the classification accuracy of Raman spectra in live animals. The 1D-CNN model yielded an 89.9% accuracy and 91.4% precision in tissue classification. Our results contribute to the field of Raman spectroscopy in cancer diagnosis, highlighting its promising role within clinical applications.

Fiber optic probe-based ATR-FTIR spectroscopy for rapid breast cancer detection: A pilot study

Breast cancer diagnosis is crucial for timely treatment and improved outcomes. This paper proposes a novel approach for rapid breast cancer diagnosis using optical fiber probe-based attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy from 750 to 4000 cm-1 . The technique enables direct analysis of tissue samples, eliminating the need for microtome sectioning and staining, thus saving time and resources. By capturing molecular fingerprint information, various machine-learning models were used to analyze the spectroscopic data to classify cancerous and non-cancerous tissues accurately. Comparing deparaffinized and paraffinized samples reveals the impact of sample preparation and experimental methods. The study demonstrates a strong correlation between the cancerous nature of a sample and its ATR-FTIR spectrum, suggesting its potential for breast cancer diagnosis (sensitivity of 74.2% and specificity of 78.3%). The proposed approach holds promise for integration into clinical operations, providing a rapid method for preliminary breast cancer diagnosis.

Needle-Based Electrical Impedance Imaging Technology for Needle Navigation

Needle insertion is a common procedure in modern healthcare practices, such as blood sampling, tissue biopsy, and cancer treatment. Various guidance systems have been developed to reduce the risk of incorrect needle positioning. While ultrasound imaging is considered the gold standard, it has limitations such as a lack of spatial resolution and subjective interpretation of 2D images. As an alternative to conventional imaging techniques, we have developed a needle-based electrical impedance imaging system. The system involves the classification of different tissue types using impedance measurements taken with a modified needle and the visualization in a MATLAB Graphical User Interface (GUI) based on the spatial sensitivity distribution of the needle. The needle was equipped with 12 stainless steel wire electrodes, and the sensitive volumes were determined using Finite Element Method (FEM) simulation. A k-Nearest Neighbors (k-NN) algorithm was used to classify different types of tissue phantoms with an average success rate of 70.56% for individual tissue phantoms. The results showed that the classification of the fat tissue phantom was the most successful (60 out of 60 attempts correct), while the success rate decreased for layered tissue structures. The measurement can be controlled in the GUI, and the identified tissues around the needle are displayed in 3D. The average latency between measurement and visualization was 112.1 ms. This work demonstrates the feasibility of using needle-based electrical impedance imaging as an alternative to conventional imaging techniques. Further improvements to the hardware and the algorithm as well as usability testing are required to evaluate the effectiveness of the needle navigation system.

Tissue Classification of Breast Cancer by Hyperspectral Unmixing

(1) Background: Assessing the resection margins during breast-conserving surgery is an important clinical need to minimize the risk of recurrent breast cancer. However, currently there is no technique that can provide real-time feedback to aid surgeons in the margin assessment. Hyperspectral imaging has the potential to overcome this problem. To classify resection margins with this technique, a tissue discrimination model should be developed, which requires a dataset with accurate ground-truth labels. However, establishing such a dataset for resection specimens is difficult. (2) Methods: In this study, we therefore propose a novel approach based on hyperspectral unmixing to determine which pixels within hyperspectral images should be assigned to the ground-truth labels from histopathology. Subsequently, we use this hyperspectral-unmixing-based approach to develop a tissue discrimination model on the presence of tumor tissue within the resection margins of ex vivo breast lumpectomy specimens. (3) Results: In total, 372 measured locations were included on the lumpectomy resection surface of 189 patients. We achieved a sensitivity of 0.94, specificity of 0.85, accuracy of 0.87, Matthew's correlation coefficient of 0.71, and area under the curve of 0.92. (4) Conclusion: Using this hyperspectral-unmixing-based approach, we demonstrated that the measured locations with hyperspectral imaging on the resection surface of lumpectomy specimens could be classified with excellent performance.

A Data-Efficient Deep Learning Strategy for Tissue Characterization via Quantitative Ultrasound: Zone Training

Deep learning (DL) powered biomedical ultrasound imaging is an emerging research field where researchers adapt the image analysis capabilities of DL algorithms to biomedical ultrasound imaging settings. A major roadblock to wider adoption of DL powered biomedical ultrasound imaging is that acquisition of large and diverse datasets is expensive in clinical settings, which is a requirement for successful DL implementation. Hence, there is a constant need for developing data-efficient DL techniques to turn DL powered biomedical ultrasound imaging into reality. In this work, we develop a data-efficient DL training strategy for classifying tissues based on the ultrasonic backscattered RF data, i.e., quantitative ultrasound (QUS), which we named zone training. In zone training, we propose to divide the complete field of view of an ultrasound image into multiple zones associated with different regions of a diffraction pattern and then, train separate DL networks for each zone. The main advantage of zone training is that it requires less training data to achieve high accuracy. In this work, three different tissue-mimicking phantoms were classified by a DL network. The results demonstrated that zone training can require a factor of 2-3 less training data in low data regime to achieve similar classification accuracies compared to a conventional training strategy.

Hyperspectral Imaging (HSI) in Brain Tumor Surgery - Evidence of Machine Learning-Based Performance

BACKGROUND

Hyperspectral imaging (HSI) has the potential to enhance surgical tissue detection and diagnostics. Definite utilization of intraoperative HSI guidance demands validated machine learning and public datasets that currently do not exist. Moreover, current imaging conventions are dispersed, and evidence-based paradigms for neurosurgical HSI have not been declared.

METHODS

We presented the rationale and a detailed clinical paradigm for establishing microneurosurgical HSI guidance. In addition, a systematic literature review was conducted to summarize the current indications and performance of neurosurgical HSI systems, with an emphasis on machine learning-based methods.

RESULTS

The published data comprised a few case series or case reports aiming to classify tissues during glioma operations. For a multi-tissue classification problem, the highest overall accuracy of 80% was obtained using deep learning. Our HSI system was capable of intraoperative data acquisition and visualization with minimal disturbance to glioma surgery.

CONCLUSIONS

In a limited number of publications, neurosurgical HSI has demonstrated unique capabilities in contrast to the established imaging techniques. Multidisciplinary work is required to establish communicable HSI standards and clinical impact. Our HSI paradigm endorses systematic intraoperative HSI data collection, which aims to facilitate the related standards, medical device regulations, and value-based medical imaging systems.

Vibro-Acoustic Sensing of Instrument Interactions as a Potential Source of Texture-Related Information in Robotic Palpation

The direct tactile assessment of surface textures during palpation is an essential component of open surgery that is impeded in minimally invasive and robot-assisted surgery. When indirectly palpating with a surgical instrument, the structural vibrations from this interaction contain tactile information that can be extracted and analysed. This study investigates the influence of the parameters contact angle α and velocity v→ on the vibro-acoustic signals from this indirect palpation. A 7-DOF robotic arm, a standard surgical instrument, and a vibration measurement system were used to palpate three different materials with varying α and v→. The signals were processed based on continuous wavelet transformation. They showed material-specific signatures in the time-frequency domain that retained their general characteristic for varying α and v→. Energy-related and statistical features were extracted, and supervised classification was performed, where the testing data comprised only signals acquired with different palpation parameters than for training data. The classifiers support vector machine and k-nearest neighbours provided 99.67% and 96.00% accuracy for the differentiation of the materials. The results indicate the robustness of the features against variations in the palpation parameters. This is a prerequisite for an application in minimally invasive surgery but needs to be confirmed in realistic experiments with biological tissues.

Diseased thyroid tissue classification in OCT images using deep learning: Towards surgical decision support

Intraoperative guidance tools for thyroid surgery based on optical coherence tomography (OCT) could aid distinguish between normal and diseased tissue. However, OCT images are difficult to interpret, thus, real-time automatic analysis could support the clinical decision-making. In this study, several deep learning models were investigated for thyroid disease classification on 2D and 3D OCT data obtained from ex vivo specimens of 22 patients undergoing surgery and diagnosed with several thyroid pathologies. Additionally, two open-access datasets were used to evaluate the custom models. On the thyroid dataset, the best performance was achieved by the 3D vision transformer model with a Matthew's correlation coefficient (MCC) of 0.79 (accuracy = 0.90) for the normal-versus-abnormal classification. On the open-access datasets, the custom models achieved the best performance (MCC > 0.88, accuracy > 0.96). Results obtained for the normal-versus-abnormal classification suggest OCT, complemented with deep learning-based analysis, as a tool for real-time automatic diseased tissue identification in thyroid surgery.

Quantitative characterization of duodenal gastrinoma autofluorescence using multiphoton microscopy

BACKGROUND

Duodenal gastrinomas (DGASTs) are neuroendocrine tumors that develop in the submucosa of the duodenum and produce the hormone gastrin. Surgical resection of DGASTs is complicated by the small size of these tumors and the tendency for them to develop diffusely in the duodenum. Endoscopic mucosal resection of DGASTs is an increasingly popular method for treating this disease due to its low complication rate but suffers from poor rates of pathologically negative margins. Multiphoton microscopy can capture high-resolution images of biological tissue with contrast generated from endogenous fluorescence (autofluorescence [AF]) through two-photon excited fluorescence (2PEF). Second harmonic generation is another popular method of generating image contrast with multiphoton microscopy (MPM) and is a light-scattering phenomenon that occurs predominantly from structures such as collagen in biological samples. Some molecules that contribute to AF change in abundance from processes related to the cancer disease process (e.g., metabolic changes, oxidative stress, and angiogenesis).

STUDY DESIGN/MATERIALS AND METHODS

MPM was used to image 12 separate patient samples of formalin-fixed and paraffin-embedded duodenal gastrinoma slides with a second-harmonic generation (SHG) channel and four 2PEF channels. The excitation and emission profiles of each 2PEF channel were tuned to capture signal dominated by distinct fluorophores with well-characterized fluorescent spectra and known connections to the physiologic changes that arise in cancerous tissue.

RESULTS

We found that there was a significant difference in the relative abundance of signal generated in the 2PEF channels for regions of DGASTs in comparison to the neighboring tissues of the duodenum. Data generated from texture feature extraction of the MPM images were used in linear discriminant analysis models to create classifiers for tumor versus all other tissue types before and after principal component analysis (PCA). PCA improved the classifier accuracy and reduced the number of features required to achieve maximum accuracy. The linear discriminant classifier after PCA distinguished between tumor and other tissue types with an accuracy of 90.6%-93.8%.

CONCLUSIONS

These results suggest that multiphoton microscopy 2PEF and SHG imaging is a promising label-free method for discriminating between DGASTs and normal duodenal tissue which has implications for future applications of in vivo assessment of resection margins with endoscopic MPM.

The neurosurgical benefit of contactless in vivo optical coherence tomography regarding residual tumor detection: A clinical study

Purpose

In brain tumor surgery, it is crucial to achieve complete tumor resection while conserving adjacent noncancerous brain tissue. Several groups have demonstrated that optical coherence tomography (OCT) has the potential of identifying tumorous brain tissue. However, there is little evidence on human in vivo application of this technology, especially regarding applicability and accuracy of residual tumor detection (RTD). In this study, we execute a systematic analysis of a microscope integrated OCT-system for this purpose.

Experimental design

Multiple 3-dimensional in vivo OCT-scans were taken at protocol-defined sites at the resection edge in 21 brain tumor patients. The system was evaluated for its intraoperative applicability. Tissue biopsies were obtained at these locations, labeled by a neuropathologist and used as ground truth for further analysis. OCT-scans were visually assessed with a qualitative classifier, optical OCT-properties were obtained and two artificial intelligence (AI)-assisted methods were used for automated scan classification. All approaches were investigated for accuracy of RTD and compared to common techniques.

Results

Visual OCT-scan classification correlated well with histopathological findings. Classification with measured OCT image-properties achieved a balanced accuracy of 85%. A neuronal network approach for scan feature recognition achieved 82% and an auto-encoder approach 85% balanced accuracy. Overall applicability showed need for improvement.

Conclusion

Contactless in vivo OCT scanning has shown to achieve high values of accuracy for RTD, supporting what has well been described for ex vivo OCT brain tumor scanning, complementing current intraoperative techniques and even exceeding them in accuracy, while not yet in applicability.

Computer-aided classification of indirect immunofluorescence patterns on esophagus and split skin for the detection of autoimmune dermatoses

Autoimmune bullous dermatoses (AIBD) are rare diseases that affect human skin and mucous membranes. Clinically, they are characterized by blister formation and/or erosions. Depending on the structures involved and the depth of blister formation, they are grouped into pemphigus diseases, pemphigoid diseases, and dermatitis herpetiformis. Classification of AIBD into their sub-entities is crucial to guide treatment decisions. One of the most sensitive screening methods for initial differentiation of AIBD is the indirect immunofluorescence (IIF) microscopy on tissue sections of monkey esophagus and primate salt-split skin, which are used to detect disease-specific autoantibodies. Interpretation of IIF patterns requires a detailed examination of the image by trained professionals automating this process is a challenging task with these highly complex tissue substrates, but offers the great advantage of an objective result. Here, we present computer-aided classification of esophagus and salt-split skin IIF images. We show how deep networks can be adapted to the specifics and challenges of IIF image analysis by incorporating segmentation of relevant regions into the prediction process, and demonstrate their high accuracy. Using this semi-automatic extension can reduce the workload of professionals when reading tissue sections in IIF testing. Furthermore, these results on highly complex tissue sections show that further integration of semi-automated workflows into the daily workflow of diagnostic laboratories is promising.

Classification of tissue biopsies by Raman spectroscopy guided by quantitative phase imaging and its application to bladder cancer

We present a multimodal label-free optical measurement approach for analyzing sliced tissue biopsies by a unique combination of quantitative phase imaging and localized Raman spectroscopy. First, label-free quantitative phase imaging of the entire unstained tissue slice is performed using automated scanning. Then, pixel-wise segmentation of the tissue layers is performed by a kernelled structural support vector machine based on Haralick texture features, which are extracted from the quantitative phase profile, and used to find the best locations for performing the label-free localized Raman measurements. We use this multimodal label-free measurement approach for segmenting the urothelium in benign and malignant bladder cancer tissues by quantitative phase imaging, followed by location-guided Raman spectroscopy measurements. We then use sparse multinomial logistic regression (SMLR) on the Raman spectroscopy measurements to classify the tissue types, demonstrating that the prior segmentation of the urothelium done by label-free quantitative phase imaging improves the Raman spectra classification accuracy from 85.7% to 94.7%.

Real-time tracking of a diffuse reflectance spectroscopy probe used to aid histological validation of margin assessment in upper gastrointestinal cancer resection surgery

SIGNIFICANCE

Diffuse reflectance spectroscopy (DRS) allows discrimination of tissue type. Its application is limited by the inability to mark the scanned tissue and the lack of real-time measurements.

AIM

This study aimed to develop a real-time tracking system to enable localization of a DRS probe to aid the classification of tumor and non-tumor tissue.

APPROACH

A green-colored marker attached to the DRS probe was detected using hue-saturation-value (HSV) segmentation. A live, augmented view of tracked optical biopsy sites was recorded in real time. Supervised classifiers were evaluated in terms of sensitivity, specificity, and overall accuracy. A developed software was used for data collection, processing, and statistical analysis.

RESULTS

The measured root mean square error (RMSE) of DRS probe tip tracking was 1.18  ±  0.58  mm and 1.05  ±  0.28  mm for the x and y dimensions, respectively. The diagnostic accuracy of the system to classify tumor and non-tumor tissue in real time was 94% for stomach and 96% for the esophagus.

CONCLUSIONS

We have successfully developed a real-time tracking and classification system for a DRS probe. When used on stomach and esophageal tissue for tumor detection, the accuracy derived demonstrates the strength and clinical value of the technique to aid margin assessment in cancer resection surgery.

Wound Size Imaging: Ready for Smart Assessment and Monitoring

Significance: We introduce and evaluate emerging devices and modalities for wound size imaging and also promising image processing tools for smart wound assessment and monitoring. Recent Advances: Some commercial devices are available for optical wound assessment but with limited possibilities compared to the power of multimodal imaging. With new low-cost devices and machine learning, wound assessment has become more robust and accurate. Wound size imaging not only provides area and volume but also the proportion of each tissue on the wound bed. Near-infrared and thermal spectral bands also enhance the classical visual assessment. Critical Issues: The ability to embed advanced imaging technology in portable devices such as smartphones and tablets with tissue analysis software tools will significantly improve wound care. As wound care and measurement are performed by nurses, the equipment needs to remain user-friendly, enable quick measurements, provide advanced monitoring, and be connected to the patient data management system. Future Directions: Combining several image modalities and machine learning, optical wound assessment will be smart enough to enable real wound monitoring, to provide clinicians with relevant indications to adapt the treatments and to improve healing rates and speed. Sharing the wound care histories of a number of patients on databases and through telemedicine practice could induce a better knowledge of the healing process and thus a better efficiency when the recorded clinical experience has been converted into knowledge through deep learning.

Toward optoacoustic sciatic nerve detection using an all-fiber interferometric-based sensor for endoscopic smart laser surgery

Objectives

Laser surgery requires efficient tissue classification to reduce the probability of undesirable or unwanted tissue damage. This study aimed to investigate acoustic shock waves (ASWs) as a means of classifying sciatic nerve tissue.

Materials and Methods

In this study, we classified sciatic nerve tissue against other tissue types—hard bone, soft bone, fat, muscle, and skin extracted from two proximal and distal fresh porcine femurs—using the ASWs generated by a laser during ablation. A nanosecond frequency‐doubled Nd:YAG laser at 532 nm was used to create 10 craters on each tissue type's surface. We used a fiber‐coupled Fabry–Pérot sensor to measure the ASWs. The spectrum's amplitude from each ASW frequency band measured was used as input for principal component analysis (PCA). PCA was combined with an artificial neural network to classify the tissue types. A confusion matrix and receiver operating characteristic (ROC) analysis was used to calculate the accuracy of the testing‐data‐based scores from the sciatic nerve and the area under the ROC curve (AUC) with a 95% confidence‐level interval.

Results

Based on the confusion matrix and ROC analysis of the model's tissue classification results (leave‐one‐out cross‐validation), nerve tissue could be classified with an average accuracy rate and AUC result of 95.78  ± 1.3% and 99.58  ± 0.6%, respectively.

Conclusion

This study demonstrates the potential of using ASWs for remote classification of nerve and other tissue types. The technique can serve as the basis of a feedback control system to detect and preserve sciatic nerves in endoscopic laser surgery.

Multispectral Imaging for Automated Tissue Identification of Normal Human Surgical Specimens

OBJECTIVE

Safe surgery requires the accurate discrimination of tissue intraoperatively. We assess the feasibility of using multispectral imaging and deep learning to enhance surgical vision by automated identification of normal human head and neck tissues.

STUDY DESIGN

Construction and feasibility testing of novel multispectral imaging system for surgery.

SETTING

Academic university hospital.

SUBJECTS AND METHODS

Multispectral images of fresh-preserved human cadaveric tissues were captured with our adapted digital operating microscope. Eleven tissue types were sampled, each sequentially exposed to 6 lighting conditions. Two convolutional neural network machine learning models were developed to classify tissues based on multispectral and white-light color images (ARRInet-M and ARRInet-W, respectively). Blinded otolaryngology residents were asked to identify tissue specimens from white-light color images, and their performance was compared with that of the ARRInet models.

RESULTS

A novel multispectral imaging system was developed with minimal adaptation to an existing digital operating microscope. With 81.8% accuracy in tissue identification of full-size images, the multispectral ARRInet-M classifier outperformed the white-light-only ARRInet-W model (45.5%) and surgical residents (69.7%). Challenges with discrimination occurred with parotid vs fat and blood vessels vs nerve.

CONCLUSIONS

A deep learning model using multispectral imaging outperformed a similar model and surgical residents using traditional white-light imaging at the task of classifying normal human head and neck tissue ex vivo. These results suggest that multispectral imaging can enhance surgical vision and augment surgeons' ability to identify tissues during a procedure.

Application of Classification Algorithms to Diffuse Reflectance Spectroscopy Measurements for Ex Vivo Characterization of Biological Tissues

Biological tissue identification in real clinical scenarios is a relevant and unsolved medical problem, particularly in the operating room. Although it could be thought that healthy tissue identification is an immediate task, in practice there are several clinical situations that greatly impede this process. For instance, it could be challenging in open surgery in complex areas, such as the neck, where different structures are quite close together, with bleeding and other artifacts affecting visual inspection. Solving this issue requires, on one hand, a high contrast noninvasive technique and, on the other hand, powerful classification algorithms. Regarding the technique, optical diffuse reflectance spectroscopy has demonstrated such capabilities in the discrimination of tumoral and healthy biological tissues. The complex signals obtained, in the form of spectra, need to be adequately computed in order to extract relevant information for discrimination. As usual, accurate discrimination relies on massive measurements, some of which serve as training sets for the classification algorithms. In this work, diffuse reflectance spectroscopy is proposed, implemented, and tested as a potential technique for healthy tissue discrimination. A specific setup is built and spectral measurements on several ex vivo porcine tissues are obtained. The massive data obtained are then analyzed for classification purposes. First of all, considerations about normalization, detrending and noise are taken into account. Dimensionality reduction and tendencies extraction are also considered. Featured spectral characteristics, principal component or linear discrimination analysis are applied, as long as classification approaches based on k-nearest neighbors (k-NN), quadratic discrimination analysis (QDA) or Naïve Bayes (NB). Relevant parameters about classification accuracy are obtained and compared, including ANOVA tests. The results show promising values of specificity and sensitivity of the technique for some classification algorithms, even over 95%, which could be relevant for clinical applications in the operating room.

Machine Learning-Based Optoacoustic Tissue Classification Method for Laser Osteotomes Using an Air-Coupled Transducer

BACKGROUND AND OBJECTIVES

Using lasers instead of mechanical tools for bone cutting holds many advantages, including functional cuts, contactless interaction, and faster wound healing. To fully exploit the benefits of lasers over conventional mechanical tools, a real-time feedback to classify tissue is proposed.

STUDY DESIGN/MATERIALS AND METHODS

In this paper, we simultaneously classified five tissue types-hard and soft bone, muscle, fat, and skin from five proximal and distal fresh porcine femurs-based on the laser-induced acoustic shock waves (ASWs) generated. For laser ablation, a nanosecond frequency-doubled Nd:YAG laser source at 532 nm and a microsecond Er:YAG laser source at 2940 nm were used to create 10 craters on the surface of each proximal and distal femur. Depending on the application, the Nd:YAG or Er:YAG can be used for bone cutting. For ASW recording, an air-coupled transducer was placed 5 cm away from the ablated spot. For tissue classification, we analyzed the measured acoustics by looking at the amplitude-frequency band of 0.11-0.27 and 0.27-0.53 MHz, which provided the least average classification error for Er:YAG and Nd:YAG, respectively. For data reduction, we used the amplitude-frequency band as an input of the principal component analysis (PCA). On the basis of PCA scores, we compared the performance of the artificial neural network (ANN), the quadratic- and Gaussian-support vector machine (SVM) to classify tissue types. A set of 14,400 data points, measured from 10 craters in four proximal and distal femurs, was used as training data, while a set of 3,600 data points from 10 craters in the remaining proximal and distal femur was considered as testing data, for each laser.

RESULTS

The ANN performed best for both lasers, with an average classification error for all tissues of 5.01 ± 5.06% and 9.12 ± 3.39%, using the Nd:YAG and Er:YAG lasers, respectively. Then, the Gaussian-SVM performed better than the quadratic SVM during the cutting with both lasers. The Gaussian-SVM yielded average classification errors of 15.17 ± 13.12% and 16.85 ± 7.59%, using the Nd:YAG and Er:YAG lasers, respectively. The worst performance was achieved with the quadratic-SVM with a classification error of 50.34 ± 35.04% and 69.96 ± 25.49%, using the Nd:YAG and Er:YAG lasers.

CONCLUSION

We foresee using the ANN to differentiate tissues in real-time during laser osteotomy. Lasers Surg. Med. © 2020 Wiley Periodicals LLC.

Reliability assessment of tissue classification algorithms for multi-center and multi-scanner data

BACKGROUND

Gray and white matter volume difference and change are important imaging markers of pathology and disease progression in neurology and psychiatry. Such measures are usually estimated from tissue segmentation maps produced by publicly available image processing pipelines. However, the reliability of the produced segmentations when using multi-center and multi-scanner data remains understudied. Here, we assess the robustness of six publicly available tissue classification pipelines across images acquired from different MR scanners and sites.

METHODS

We used 90 T1-weighted images of a single individual, scanned in 73 sessions across 27 different sites to assess the robustness of the tissue classification tools. Variability in Dice similarity index values and tissue volumes was assessed for Atropos, BISON, Classify_Clean, FAST, FreeSurfer, and SPM12.

RESULTS

BISON had the highest overall Dice coefficient for GM, followed by SPM12 and Atropos; while Atropos had the highest overall Dice coefficient for WM, followed by BISON and SPM12. BISON had the lowest overall variability in its volumetric estimates, followed by FreeSurfer, and SPM12. All methods also had significant differences between some of their estimates across different scanner manufacturers (e.g. BISON had significantly higher GM estimates and correspondingly lower WM estimates for GE scans compared to Philips and Siemens), and different signal-to-noise ratio (SNR) levels (e.g. FAST and FreeSurfer had significantly higher WM volume estimates for high versus medium and low SNR tertiles as well as correspondingly lower GM volume estimates).

CONCLUSIONS

Our comparisons provide a benchmark on the reliability of the publicly used tissue classification techniques and the amount of variability that can be expected when using large multi-center and multi-scanner databases.

Brain and lesion segmentation in multiple sclerosis using fully convolutional neural networks: A large-scale study

OBJECTIVE

To investigate the performance of deep learning (DL) based on fully convolutional neural network (FCNN) in segmenting brain tissues in a large cohort of multiple sclerosis (MS) patients.

METHODS

We developed a FCNN model to segment brain tissues, including T2-hyperintense MS lesions. The training, validation, and testing of FCNN were based on ~1000 magnetic resonance imaging (MRI) datasets acquired on relapsing-remitting MS patients, as a part of a phase 3 randomized clinical trial. Multimodal MRI data (dual-echo, FLAIR, and T1-weighted images) served as input to the network. Expert validated segmentation was used as the target for training the FCNN. We cross-validated our results using the leave-one-center-out approach.

RESULTS

We observed a high average (95% confidence limits) Dice similarity coefficient for all the segmented tissues: 0.95 (0.92-0.98) for white matter, 0.96 (0.93-0.98) for gray matter, 0.99 (0.98-0.99) for cerebrospinal fluid, and 0.82 (0.63-1.0) for T2 lesions. High correlations between the DL segmented tissue volumes and ground truth were observed (R² > 0.92 for all tissues). The cross validation showed consistent results across the centers for all tissues.

CONCLUSION

The results from this large-scale study suggest that deep FCNN can automatically segment MS brain tissues, including lesions, with high accuracy.

Brain Segmentation From Computed Tomography of Healthy Aging and Geriatric Concussion at Variable Spatial Resolutions

When properly implemented and processed, anatomic T ₁-weighted magnetic resonance imaging (MRI) can be ideal for the noninvasive quantification of white matter (WM) and gray matter (GM) in the living human brain. Although MRI is more suitable for distinguishing GM from WM than computed tomography (CT), the growing clinical use of the latter technique has renewed interest in head CT segmentation. Such interest is particularly strong in settings where MRI is unavailable, logistically unfeasible or prohibitively expensive. Nevertheless, whereas MRI segmentation is a sophisticated and technically-mature research field, the task of automatically classifying soft brain tissues from CT remains largely unexplored. Furthermore, brain segmentation methods for MRI hold considerable potential for adaptation and application to CT image processing. Here we demonstrate this by combining probabilistic, atlas-based classification with topologically-constrained tissue boundary refinement to delineate WM, GM and cerebrospinal fluid (CSF) from head CT images. The feasibility and utility of this approach are revealed by comparison of MRI-only vs. CT-only segmentations in geriatric concussion victims with both MRI and CT scans. Comparison of the two segmentations yields mean Sørensen-Dice coefficients of 85.5 ± 4.6% (WM), 86.7 ± 5.6% (GM) and 91.3 ± 2.8% (CSF), as well as average Hausdorff distances of 3.76 ± 1.85 mm (WM), 3.43 ± 1.53 mm (GM) and 2.46 ± 1.27 mm (CSF). Bootstrapping results suggest that the segmentation approach is sensitive enough to yield WM, GM and CSF volume estimates within ~5%, ~4%, and ~3% of their MRI-based estimates, respectively. To our knowledge, this is the first 3D segmentation approach for CT to undergo rigorous within-subject comparison with high-resolution MRI. Results suggest that (1) standard-quality CT allows WM/GM/CSF segmentation with reasonable accuracy, and that (2) the task of soft brain tissue classification from CT merits further attention from neuroimaging researchers.

A Computational Method for Classifying Different Human Tissues with Quantitatively Tissue-Specific Expressed Genes

Tissue-specific gene expression has long been recognized as a crucial key for understanding tissue development and function. Efforts have been made in the past decade to identify tissue-specific expression profiles, such as the Human Proteome Atlas and FANTOM5. However, these studies mainly focused on "qualitatively tissue-specific expressed genes" which are highly enriched in one or a group of tissues but paid less attention to "quantitatively tissue-specific expressed genes", which are expressed in all or most tissues but with differential expression levels. In this study, we applied machine learning algorithms to build a computational method for identifying "quantitatively tissue-specific expressed genes" capable of distinguishing 25 human tissues from their expression patterns. Our results uncovered the expression of 432 genes as optimal features for tissue classification, which were obtained with a Matthews Correlation Coefficient (MCC) of more than 0.99 yielded by a support vector machine (SVM). This constructed model was superior to the SVM model using tissue enriched genes and yielded MCC of 0.985 on an independent test dataset, indicating its good generalization ability. These 432 genes were proven to be widely expressed in multiple tissues and a literature review of the top 23 genes found that most of them support their discriminating powers. As a complement to previous studies, our discovery of these quantitatively tissue-specific genes provides insights into the detailed understanding of tissue development and function.

Intraoperative hyperspectral determination of human tissue properties

We address the automatic differentiation of human tissue using multispectral imaging with promising potential for automatic visualization during surgery. Currently, tissue types have to be continuously differentiated based on the surgeon's knowledge only. Further, automatic methods based on optical in vivo properties of human tissue do not yet exist, as these properties have not been sufficiently examined. To overcome this, we developed a hyperspectral camera setup to monitor the different optical behavior of tissue types in vivo. The aim of this work is to collect and analyze these behaviors to open up optical opportunities during surgery. Our setup uses a digital camera and several bandpass filters in front of the light source to illuminate different tissue types with 16 specific wavelength ranges. We analyzed the different intensities of eight healthy tissue types over the visible spectrum (400 to 700 nm). Using our setup and sophisticated postprocessing in order to handle motion during capturing, we are able to find tissue characteristics not visible for the human eye to differentiate tissue types in the 16-dimensional wavelength domain. Our analysis shows that this approach has the potential to support the surgeon's decisions during treatment.

Confident texture-based laryngeal tissue classification for early stage diagnosis support

Early stage diagnosis of laryngeal squamous cell carcinoma (SCC) is of primary importance for lowering patient mortality or after treatment morbidity. Despite the challenges in diagnosis reported in the clinical literature, few efforts have been invested in computer-assisted diagnosis. The objective of this paper is to investigate the use of texture-based machine-learning algorithms for early stage cancerous laryngeal tissue classification. To estimate the classification reliability, a measure of confidence is also exploited. From the endoscopic videos of 33 patients affected by SCC, a well-balanced dataset of 1320 patches, relative to four laryngeal tissue classes, was extracted. With the best performing feature, the achieved median classification recall was 93% [interquartile range [Formula: see text]]. When excluding low-confidence patches, the achieved median recall was increased to 98% ([Formula: see text]), proving the high reliability of the proposed approach. This research represents an important advancement in the state-of-the-art computer-assisted laryngeal diagnosis, and the results are a promising step toward a helpful endoscope-integrated processing system to support early stage diagnosis.

Visualization and tissue classification of human breast cancer images using ultrahigh-resolution OCT

BACKGROUND AND OBJECTIVE

Breast cancer is one of the most common cancers, and recognized as the third leading cause of mortality in women. Optical coherence tomography (OCT) enables three dimensional visualization of biological tissue with micrometer level resolution at high speed, and can play an important role in early diagnosis and treatment guidance of breast cancer. In particular, ultra-high resolution (UHR) OCT provides images with better histological correlation. This paper compared UHR OCT performance with standard OCT in breast cancer imaging qualitatively and quantitatively. Automatic tissue classification algorithms were used to automatically detect invasive ductal carcinoma in ex vivo human breast tissue.

STUDY DESIGN/MATERIALS AND METHODS

Human breast tissues, including non-neoplastic/normal tissues from breast reduction and tumor samples from mastectomy specimens, were excised from patients at Columbia University Medical Center. The tissue specimens were imaged by two spectral domain OCT systems at different wavelengths: a home-built ultra-high resolution (UHR) OCT system at 800 nm (measured as 2.72 μm axial and 5.52 μm lateral) and a commercial OCT system at 1,300 nm with standard resolution (measured as 6.5 μm axial and 15 μm lateral), and their imaging performances were analyzed qualitatively. Using regional features derived from OCT images produced by the two systems, we developed an automated classification algorithm based on relevance vector machine (RVM) to differentiate hollow-structured adipose tissue against solid tissue. We further developed B-scan based features for RVM to classify invasive ductal carcinoma (IDC) against normal fibrous stroma tissue among OCT datasets produced by the two systems. For adipose classification, 32 UHR OCT B-scans from 9 normal specimens, and 28 standard OCT B-scans from 6 normal and 4 IDC specimens were employed. For IDC classification, 152 UHR OCT B-scans from 6 normal and 13 IDC specimens, and 104 standard OCT B-scans from 5 normal and 8 IDC specimens were employed.

RESULTS

We have demonstrated that UHR OCT images can produce images with better feature delineation compared with images produced by 1,300 nm OCT system. UHR OCT images of a variety of tissue types found in human breast tissue were presented. With a limited number of datasets, we showed that both OCT systems can achieve a good accuracy in identifying adipose tissue. Classification in UHR OCT images achieved higher sensitivity (94%) and specificity (93%) of adipose tissue than the sensitivity (91%) and specificity (76%) in 1,300 nm OCT images. In IDC classification, similarly, we achieved better results with UHR OCT images, featured an overall accuracy of 84%, sensitivity of 89% and specificity of 71% in this preliminary study.

CONCLUSION

In this study, we provided UHR OCT images of different normal and malignant breast tissue types, and qualitatively and quantitatively studied the texture and optical features from OCT images of human breast tissue at different resolutions. We developed an automated approach to differentiate adipose tissue, fibrous stroma, and IDC within human breast tissues. Our work may open the door toward automatic intraoperative OCT evaluation of early-stage breast cancer. Lasers Surg. Med. 49:258-269, 2017. © 2017 Wiley Periodicals, Inc.

Tissue classification for laparoscopic image understanding based on multispectral texture analysis

Intraoperative tissue classification is one of the prerequisites for providing context-aware visualization in computer-assisted minimally invasive surgeries. As many anatomical structures are difficult to differentiate in conventional RGB medical images, we propose a classification method based on multispectral image patches. In a comprehensive ex vivo study through statistical analysis, we show that (1) multispectral imaging data are superior to RGB data for organ tissue classification when used in conjunction with widely applied feature descriptors and (2) combining the tissue texture with the reflectance spectrum improves the classification performance. The classifier reaches an accuracy of 98.4% on our dataset. Multispectral tissue analysis could thus evolve as a key enabling technique in computer-assisted laparoscopy.

Neonatal Brain Tissue Classification with Morphological Adaptation and Unified Segmentation

Measuring the distribution of brain tissue types (tissue classification) in neonates is necessary for studying typical and atypical brain development, such as that associated with preterm birth, and may provide biomarkers for neurodevelopmental outcomes. Compared with magnetic resonance images of adults, neonatal images present specific challenges that require the development of specialized, population-specific methods. This paper introduces MANTiS (Morphologically Adaptive Neonatal Tissue Segmentation), which extends the unified segmentation approach to tissue classification implemented in Statistical Parametric Mapping (SPM) software to neonates. MANTiS utilizes a combination of unified segmentation, template adaptation via morphological segmentation tools and topological filtering, to segment the neonatal brain into eight tissue classes: cortical gray matter, white matter, deep nuclear gray matter, cerebellum, brainstem, cerebrospinal fluid (CSF), hippocampus and amygdala. We evaluated the performance of MANTiS using two independent datasets. The first dataset, provided by the NeoBrainS12 challenge, consisted of coronal T₂-weighted images of preterm infants (born ≤30 weeks' gestation) acquired at 30 weeks' corrected gestational age (n = 5), coronal T₂-weighted images of preterm infants acquired at 40 weeks' corrected gestational age (n = 5) and axial T₂-weighted images of preterm infants acquired at 40 weeks' corrected gestational age (n = 5). The second dataset, provided by the Washington University NeuroDevelopmental Research (WUNDeR) group, consisted of T₂-weighted images of preterm infants (born <30 weeks' gestation) acquired shortly after birth (n = 12), preterm infants acquired at term-equivalent age (n = 12), and healthy term-born infants (born ≥38 weeks' gestation) acquired within the first 9 days of life (n = 12). For the NeoBrainS12 dataset, mean Dice scores comparing MANTiS with manual segmentations were all above 0.7, except for the cortical gray matter for coronal images acquired at 30 weeks. This demonstrates that MANTiS' performance is competitive with existing techniques. For the WUNDeR dataset, mean Dice scores comparing MANTiS with manually edited segmentations demonstrated good agreement, where all scores were above 0.75, except for the hippocampus and amygdala. The results show that MANTiS is able to segment neonatal brain tissues well, even in images that have brain abnormalities common in preterm infants. MANTiS is available for download as an SPM toolbox from http://developmentalimagingmcri.github.io/mantis.

Multispectral tissue characterization for intestinal anastomosis optimization

Intestinal anastomosis is a surgical procedure that restores bowel continuity after surgical resection to treat intestinal malignancy, inflammation, or obstruction. Despite the routine nature of intestinal anastomosis procedures, the rate of complications is high. Standard visual inspection cannot distinguish the tissue subsurface and small changes in spectral characteristics of the tissue, so existing tissue anastomosis techniques that rely on human vision to guide suturing could lead to problems such as bleeding and leakage from suturing sites. We present a proof-of-concept study using a portable multispectral imaging (MSI) platform for tissue characterization and preoperative surgical planning in intestinal anastomosis. The platform is composed of a fiber ring light-guided MSI system coupled with polarizers and image analysis software. The system is tested on ex vivo porcine intestine tissue, and we demonstrate the feasibility of identifying optimal regions for suture placement.

Color correction for automatic fibrosis quantification in liver biopsy specimens

Context:

For a precise and objective quantification of liver fibrosis, quantitative evaluations through image analysis have been utilized. However, manual operations are required in most cases for extracting fiber areas because of color variation included in digital pathology images.

Aims:

The purpose of this research is to propose a color correction method for whole slide images (WSIs) of Elastica van Gieson (EVG) stained liver biopsy tissue specimens and to realize automated operation of image analysis for fibrosis quantification.

Materials and Methods:

Our experimental dataset consisted of 38 WSIs of liver biopsy specimens collected from 38 chronic viral hepatitis patients from multiple medical facilities, stained with EVG and scanned at ×20 using a Nano Zoomer 2.0 HT (Hamamatsu Photonics K.K., Hamamatsu, Japan). Color correction was performed by modifying the color distribution of a target WSI so as to fit to the reference, where the color distribution was modeled by a set of two triangle pyramids. Using color corrected WSIs; fibrosis quantification was performed based on tissue classification analysis.

Statistical Analysis Used:

Spearman's rank correlation coefficients were calculated between liver stiffness measured by transient elastography and median area ratio of collagen fibers calculated based on tissue classification results.

Results:

Statistical analysis results showed a significant correlation r = 0.61-0.68 even when tissue classifiers were trained by using a subset of WSIs, while the correlation coefficients were reduced to r = 0.40-0.50 without color correction.

Conclusions:

Fibrosis quantification accompanied with the proposed color correction method could provide an objective evaluation tool for liver fibrosis, which complements semi-quantitative histologic evaluation systems.

Robust multi-site MR data processing: iterative optimization of bias correction, tissue classification, and registration

A robust multi-modal tool, for automated registration, bias correction, and tissue classification, has been implemented for large-scale heterogeneous multi-site longitudinal MR data analysis. This work focused on improving the an iterative optimization framework between bias-correction, registration, and tissue classification inspired from previous work. The primary contributions are robustness improvements from incorporation of following four elements: (1) utilize multi-modal and repeated scans, (2) incorporate high-deformable registration, (3) use extended set of tissue definitions, and (4) use of multi-modal aware intensity-context priors. The benefits of these enhancements were investigated by a series of experiments with both simulated brain data set (BrainWeb) and by applying to highly-heterogeneous data from a 32 site imaging study with quality assessments through the expert visual inspection. The implementation of this tool is tailored for, but not limited to, large-scale data processing with great data variation with a flexible interface. In this paper, we describe enhancements to a joint registration, bias correction, and the tissue classification, that improve the generalizability and robustness for processing multi-modal longitudinal MR scans collected at multi-sites. The tool was evaluated by using both simulated and simulated and human subject MRI images. With these enhancements, the results showed improved robustness for large-scale heterogeneous MRI processing.

Cortical surface reconstruction via unified Reeb analysis of geometric and topological outliers in magnetic resonance images

In this paper we present a novel system for the automated reconstruction of cortical surfaces from T1-weighted magnetic resonance images. At the core of our system is a unified Reeb analysis framework for the detection and removal of geometric and topological outliers on tissue boundaries. Using intrinsic Reeb analysis, our system can pinpoint the location of spurious branches and topological outliers, and correct them with localized filtering using information from both image intensity distributions and geometric regularity. In this system, we have also developed enhanced tissue classification with Hessian features for improved robustness to image inhomogeneity, and adaptive interpolation to achieve sub-voxel accuracy in reconstructed surfaces. By integrating these novel developments, we have a system that can automatically reconstruct cortical surfaces with improved quality and dramatically reduced computational cost as compared with the popular FreeSurfer software. In our experiments, we demonstrate on 40 simulated MR images and the MR images of 200 subjects from two databases: the Alzheimer’s Disease Neuroimaging Initiative (ADNI) and International Consortium of Brain Mapping (ICBM), the robustness of our method in large scale studies. In comparisons with FreeSurfer, we show that our system is able to generate surfaces that better represent cortical anatomy and produce thickness features with higher statistical power in population studies.

The AEO, an Ontology of Anatomical Entities for Classifying Animal Tissues and Organs

This paper describes the AEO, an ontology of anatomical entities that expands the common anatomy reference ontology (CARO) and whose major novel feature is a type hierarchy of ~160 anatomical terms. The breadth of the AEO is wider than CARO as it includes both developmental and gender-specific classes, while the granularity of the AEO terms is at a level adequate to classify simple-tissues (~70 classes) characterized by their containing a predominantly single cell-type. For convenience and to facilitate interoperability, the AEO contains an abbreviated version of the ontology of cell-types (~100 classes) that is linked to these simple-tissue types. The AEO was initially based on an analysis of a broad range of animal anatomy ontologies and then upgraded as it was used to classify the ~2500 concepts in a new version of the ontology of human developmental anatomy (www.obofoundry.org/), a process that led to significant improvements in its structure and content, albeit with a possible focus on mammalian embryos. The AEO is intended to provide the formal classification expected in contemporary ontologies as well as capturing knowledge about anatomical structures not currently included in anatomical ontologies. The AEO may thus be useful in increasing the amount of tissue and cell-type knowledge in other anatomy ontologies, facilitating annotation of tissues that share common features, and enabling interoperability across anatomy ontologies. The AEO can be downloaded from http://www.obofoundry.org/.

Consistent segmentation using a Rician classifier

Several popular classification algorithms used to segment magnetic resonance brain images assume that the image intensities, or log-transformed intensities, satisfy a finite Gaussian mixture model. In these methods, the parameters of the mixture model are estimated and the posterior probabilities for each tissue class are used directly as soft segmentations or combined to form a hard segmentation. It is suggested and shown in this paper that a Rician mixture model fits the observed data better than a Gaussian model. Accordingly, a Rician mixture model is formulated and used within an expectation maximization (EM) framework to yield a new tissue classification algorithm called Rician Classifier using EM (RiCE). It is shown using both simulated and real data that RiCE yields comparable or better performance to that of algorithms based on the finite Gaussian mixture model. As well, we show that RiCE yields more consistent segmentation results when used on images of the same individual acquired with different T1-weighted pulse sequences. Therefore, RiCE has the potential to stabilize segmentation results in brain studies involving heterogeneous acquisition sources as is typically found in both multi-center and longitudinal studies.

Automatic segmentation and quantitative analysis of the articular cartilages from magnetic resonance images of the knee

In this paper, we present a segmentation scheme that automatically and accurately segments all the cartilages from magnetic resonance (MR) images of nonpathological knees. Our scheme involves the automatic segmentation of the bones using a three-dimensional active shape model, the extraction of the expected bone-cartilage interface (BCI), and cartilage segmentation from the BCI using a deformable model that utilizes localization, patient specific tissue estimation and a model of the thickness variation. The accuracy of this scheme was experimentally validated using leave one out experiments on a database of fat suppressed spoiled gradient recall MR images. The scheme was compared to three state of the art approaches, tissue classification, a modified semi-automatic watershed algorithm and nonrigid registration (B-spline based free form deformation). Our scheme obtained an average Dice similarity coefficient (DSC) of (0.83, 0.83, 0.85) for the (patellar, tibial, femoral) cartilages, while (0.82, 0.81, 0.86) was obtained with a tissue classifier and (0.73, 0.79, 0.76) was obtained with nonrigid registration. The average DSC obtained for all the cartilages using a semi-automatic watershed algorithm (0.90) was slightly higher than our approach (0.89), however unlike this approach we segment each cartilage as a separate object. The effectiveness of our approach for quantitative analysis was evaluated using volume and thickness measures with a median volume difference error of (5.92, 4.65, 5.69) and absolute Laplacian thickness difference of (0.13, 0.24, 0.12) mm.

Artificial neural networks as supervised techniques for FT-IR microspectroscopic imaging

In this report the applicability of an improved method of image segmentation of infrared microspectroscopic data from histological specimens is demonstrated. Fourier transform infrared (FT-IR) microspectroscopy was used to record hyperspectral data sets from human colorectal adenocarcinomas and to build up a database of spatially resolved tissue spectra. This database of colon microspectra comprised 4120 high-quality FT-IR point spectra from 28 patient samples and 12 different histological structures. The spectral information contained in the database was employed to teach and validate multilayer perceptron artificial neural network (MLP-ANN) models. These classification models were then employed for database analysis and utilised to produce false colour images from complete tissue maps of FT-IR microspectra. An important aspect of this study was also to demonstrate how the diagnostic sensitivity and specificity can be specifically optimised. An example is given which shows that changes of the number of teaching patterns per class can be used to modify these two interrelated test parameters. The definition of ANN topology turned out to be crucial to achieve a high degree of correspondence between the gold standard of histopathology and IR spectroscopy. Particularly, a hierarchical scheme of ANN classification proved to be superior for the reliable classification of tissue spectra. It was found that unsupervised methods of clustering, specifically agglomerative hierarchical clustering (AHC), were helpful in the initial phases of model generation. Optimal classification results could be achieved if the class definitions for the ANNs were carried out by considering the classification information provided by cluster analysis.

A feasibility study of optical coherence tomography for guiding deep brain probes

Deep brain simulation (DBS) is effective for the treatment of various diseases including Parkinson's disease and essential tremor. However, anatomical targeting combined with microelectrode mapping of the region requires significant surgical time. Also, the fine-tipped microelectrode imposes a risk of hemorrhage in the event that the trajectory intersects subcortical vessels. To reduce the operation time and the risk of hemorrhage, we propose to use optical coherence tomography (OCT) to guide the insertion of the DBS probe. We conducted in vitro experiments in the rat brain to study the feasibility of this application. The result shows that OCT is able to differentiate structures in the rat brain. White matter tends to have higher peak reflectivity and steeper attenuation rate compared to gray matter. This structural information may help guide DBS probe advance and electrical measurements.