Bronchoscopy with and without needle-based confocal laser endomicroscopy for peripheral lung nodule diagnosis: protocol for a multicentre randomised controlled trial (CLEVER trial).

INTRODUCTION: Despite many technological advances, the diagnostic yield of bronchoscopic peripheral lung nodule analysis remains limited due to frequent mispositioning. Needle-based confocal laser endomicroscopy (nCLE) enables real-time microscopic feedback on needle positioning, potentially improving the sampling location and diagnostic yield. Previous studies have defined and validated nCLE criteria for malignancy, airway and lung parenchyma. Larger studies demonstrating the effect of nCLE on diagnostic yield are lacking. We aim to investigate if nCLE-imaging integrated with conventional bronchoscopy results in a higher diagnostic yield compared with conventional bronchoscopy without nCLE. METHODS AND ANALYSIS: This is a parallel-group randomised controlled trial. Recruitment is performed at pulmonology outpatient clinics in universities and general hospitals in six different European countries and one hospital in the USA. Consecutive patients with a for malignancy suspected peripheral lung nodule (10-30 mm) with an indication for diagnostic bronchoscopy will be screened, and 208 patients will be included. Web-based randomisation (1:1) between the two procedures will be performed. The primary outcome is diagnostic yield. Secondary outcomes include diagnostic sensitivity for malignancy, needle repositionings, procedure and fluoroscopy duration, and complications. Pathologists will be blinded to procedure type; patients and endoscopists will not. ETHICS AND DISSEMINATION: Primary approval by the Ethics Committee of the Amsterdam University Medical Center. Dissemination involves publication in a peer-reviewed journal. SUPPORT: Financial and material support from Mauna Kea Technologies. TRIAL REGISTRATION NUMBER: NCT06079970.

Advanced optical imaging techniques for bladder cancer detection and diagnosis: a systematic review.

OBJECTIVES: To systematically assess the current available literature concerning advanced optical imaging methods for the detection and diagnosis of bladder cancer (BCa), focusing particularly on the sensitivity and specificity of these techniques. METHODS: First a scoping search was performed to identify all available optical techniques for BCa detection and diagnosis. The optical imaging techniques used for detecting BCa are: the Storz professional image enhancement system (IMAGE1 S), narrow-band imaging (NBI), photoacoustic imaging (PAI), autofluorescence imaging (AFI), photodynamic diagnosis (PDD), and scanning fibre endoscopy (SFE). The staging and grading techniques for BCa are: optical coherence tomography (OCT), confocal laser endomicroscopy (CLE), Raman spectroscopy, endocytoscopy, and non-linear optical microscopy (NLO). Then a systematic literature search was conducted using MEDLINE, EMBASE and Web of Science from inception to 21 November 2023. Articles were screened and selected by two independent reviewers. Inclusion criteria were: reporting on both the sensitivity and specificity of a particular technique and comparison to histopathology, and in the case of a detection technique comparison to white light cystoscopy (WLC). RESULTS: Out of 6707 articles, 189 underwent full-text review, resulting in 52 inclusions. No articles met criteria for IMAGE1 S, PAI, SFE, Raman spectroscopy, and endocytoscopy. All detection techniques showed higher sensitivity than WLC, with NBI leading (87.8-100%). Overall, detection technique specificity was comparable to WLC, with PDD being most specific (23.3-100%). CLE and OCT varied in sensitivity and specificity, with OCT showing higher specificity for BCa diagnosis, notably for carcinoma in situ (97-99%) compared to CLE (62.5-81.3%). NLO demonstrated high sensitivity and specificity (90-97% and 77-100%, respectively) based on limited data from two small ex vivo studies. CONCLUSIONS: Optical techniques with the most potential are PDD for detecting and OCT for staging and grading BCa. Further research is crucial to validate their integration into routine practice and explore the value of other imaging techniques.

Reliable Visualization of the Treatment Effect of Transperineal Focal Laser Ablation in Prostate Cancer Patients by Magnetic Resonance Imaging and Contrast-enhanced Ultrasound Imaging.

Background: Transperineal focal laser ablation (TPLA) treatment for prostate cancer (PCa) is an experimental focal ablative therapy modality with low morbidity. However, a dosimetry model for TPLA is lacking. Objective: To determine (1) the three-dimensional (3D) histologically defined ablation zone of single- and multifiber TPLA treatment for PCa correlated with magnetic resonance imaging (MRI) and contrast-enhanced ultrasound (CEUS) and (2) a reliable imaging modality of ablation zone volumetry. Design setting and participants: This was a prospective, multicenter, and interventional phase I/II pilot study with an ablate-and-resect design. TPLA was performed in 12 patients with localized prostate cancer divided over four treatment regimens to evaluate potential variation in outcomes. Intervention: TPLA was performed approximately 4 wk prior to robot-assisted radical prostatectomy (RARP) in a daycare setting using local anesthesia. Outcome measurements and statistical analysis: Four weeks after TPLA, ablation zone volumetry was determined on prostate MRI and CEUS by delineation and segmentation into 3D models and correlated with whole-mount RARP histology using the Pearson correlation index. Results and limitations: Twelve office-based TPLA procedures were performed successfully under continuous transrectal ultrasound guidance using local perineal anesthesia. No serious adverse events occurred. A qualitative analysis showed a clear demarcation of the ablation zone on T2-weighted MRI, dynamic contrast-enhanced MRI, and CEUS. On pathological evaluation, no remnant cancer was observed within the ablation zone. Ablation zone volumetry on CEUS and T2-weighted MRI compared with histology had a Pearson correlation index of r = 0.94 (95% confidence interval [CI] 0.74-0.99, p < 0.001) and r = 0.93 (95% CI 0.73-0.98, p < 0.001), respectively. Conclusions: CEUS and prostate MRI could reliably visualize TPLA ablative effects after minimally invasive PCa treatment with a high concordance with histopathological findings and showed no remnant cancer. Patient summary: The treatment effects of a novel minimally invasive ablation therapy device can reliably be visualized with radiological examinations. These results will improve planning and performance of future procedures.

Image-guided in-Vivo Needle-Based Confocal Laser Endomicroscopy in the Prostate: Safety and Feasibility Study in 2 Patients.

Purpose: To assess the safety and technical feasibility of in-vivo needle-based forward-looking confocal laser endomicroscopy in prostate tissue. Methods: For this feasibility study, 2 patients with a suspicion of prostate cancer underwent transperineal needle-based confocal laser endomicroscopy during ultrasound-guided transperineal template mapping biopsies. After intravenous administration of fluorescein, needle-based confocal laser endomicroscopy imaging was performed with a forward-looking probe (outer diameter 0.9 mm) in 2 trajectories during a manual push-forward and pullback motion. A biopsy was taken in a coregistered parallel adjacent trajectory to the confocal laser endomicroscopy trajectory for histopathologic comparison. Peri- and postprocedural adverse events, confocal laser endomicroscopy device malfunction and procedural failures were recorded. Needle-based confocal laser endomicroscopy image quality assessment, image interpretation, and histology were performed by an experienced confocal laser endomicroscopy rater and uro-pathologist, blinded to any additional information. Results: In both patients, no peri- and post-procedural adverse events were reported following needle-based confocal laser endomicroscopy. No confocal laser endomicroscopy device malfunction nor procedural failures were reported. Within 1.5 min after intravenous administration of fluorescein, needle-based confocal laser endomicroscopy image quality was sufficient for interpretation for at least 14 min, yielding more than 5000 confocal laser endomicroscopy frames per patient. The pullback confocal laser endomicroscopy recordings and most of the push-forward recordings almost only visualized erythrocytes, being classified as non-representative. During the push-forward recordings, prostate tissue was occasionally visualized in single frames, insufficient for histopathologic comparison. Prostate carcinoma was identified by biopsy in one patient (Gleason score 4 + 3 = 7, >50%), while the biopsy from the other patient showed no malignancy. Conclusion: Needle-based confocal laser endomicroscopy imaging of in-vivo prostate tissue with a forward-looking confocal laser endomicroscopy probe is safe without device malfunctions or procedural failures. Needle-based confocal laser endomicroscopy is technically feasible, but the acquired confocal laser endomicroscopy datasets are non-representative. The confocal laser endomicroscopy images' non-representative nature is possibly caused by bleeding artifacts, movement artifacts and a lack of contact time with the tissue of interest. A different confocal laser endomicroscopy probe or procedure might yield representative images of prostatic tissue.

How Reliable Is Endoscopic Stone Recognition? A Comparison Between Visual Stone Identification and Formal Stone Analysis.

Objective: To assess the diagnostic accuracy and intra-observer agreement of endoscopic stone recognition (ESR) compared with formal stone analysis. Introduction: Stone analysis is a corner stone in the prevention of stone recurrence. Although X-ray diffraction (XRD) and infrared spectroscopy are the recommended techniques for reliable formal stone analysis, this is not always possible, and the process takes time and is costly. ESR could be an alternative, as it would give immediate information on stone composition. Materials and Methods: Fifteen endourologists predicted stone composition based on 100 videos from ureterorenoscopy. Diagnostic accuracy was evaluated by comparing the prediction from visual assessment with stone analysis by XRD. After 30 days, the videos were reviewed again in a random order to assess intra-observer agreement. Results: The median diagnostic accuracy for calcium oxalate monohydrate was 54% in questionnaire 1 (Q1) and 59% in questionnaire 2 (Q2), whereas calcium oxalate dihydrate had a median diagnostic accuracy of 75% in Q1 and 50% in Q2. The diagnostic accuracy for calcium hydroxyphosphate was 10% in Q1 and 13% in Q2. The median diagnostic accuracy for calcium hydrogen phosphate dihydrate and calcium magnesium phosphate was 0% in both questionnaires. The median diagnostic accuracy for magnesium ammonium phosphate was 20% in Q1 and 40% in Q2. The median diagnostic accuracy for uric acid was 22% in both questionnaires. Finally, there was a diagnostic accuracy of 60% in Q1 and 80% in Q2 for cystine. The intra-observer agreement ranged between 45% and 72%. Conclusion: Diagnostic accuracy of ESR is limited and intra-observer agreement is below the threshold of acceptable agreement.

Safety and Feasibility of Soractelite Transperineal Focal Laser Ablation for Prostate Cancer and Short-term Quality of Life Analysis from a Multicenter Pilot Study.

Background: Soractelite™ transperineal focal laser ablation (TPLA) for the treatment of localized prostate cancer (PCa) using the Echolaser® system is a novel minimally invasive technique that has the potential to induce tissue ablation, while reducing treatment-related morbidity, when compared with robot-assisted radical prostatectomy (RARP) and radiotherapy. Objective: To determine the short-term safety and feasibility of single or multifiber TPLA, its functional outcomes, and quality of life (QoL). Design setting and participants: TPLA was performed in 12 patients, consecutively assigned to four treatment regimens, with localized PCa who were scheduled for RARP ("ablate and resect design"). The treatment regimens were as follows: (1) a single fiber at 3 W, (2) two fibers at 5 mm distance at 3 W, (3) two fibers at 10 mm distance at 3 W, and (4) a single fiber at 5 W. TPLA was scheduled 4 wk prior to RARP. Intervention: TPLA using the Echolaser® system under local anesthesia at the outpatient clinic. Outcome measurements and statistical analysis: Safety and feasibility were determined by the assessment of device-related peri- and postoperative adverse events (AEs), and length of hospital stay. Functional outcomes and QoL were measured using validated questionnaires. Feasibility of RARP was assessed by a questionnaire for the urologist. Results and limitations: Patients were dismissed after a median (interquartile range) hospital admission of 3.25 (1.25) h. No device-related AEs occurred. AEs that occurred were mostly related to lower urinary tract symptoms and were mild (grade 1-2). Most AEs resolved within 1 wk. A QoL analysis showed no significant differences for all treatment regimens. Functional outcomes remained unchanged, except for erectile function after 1 wk, which returned to baseline after 4 wk. TPLA treatment did not compromise RARP, based on the questionnaires. Conclusions: TPLA for the treatment of PCa at the outpatient clinic appears to be safe and feasible with good short-term QoL and functional outcomes; oncological results are awaited. Patient summary: Focal treatment of localized prostate cancer can safely be performed in a daycare setting using a new technique, based on laser ablation, without compromising quality of life.

Near-Infrared Fluorescence with Indocyanine Green to Assess Bone Perfusion: A Systematic Review.

Background: Adequate perfusion of a bone flap is essential for successful reconstruction of osseous defects. Unfortunately, complications related to inadequate bone perfusion are common. Near-infrared fluorescence (NIRF) imaging enables intraoperative visualization of perfusion. NIRF has been investigated in reconstructive surgery to aid the surgeon in clinical perioperative assessment of soft tissue perfusion. However, little is known on the beneficial use of NIRF to assess bone perfusion. Therefore, the aim of this review was to search for studies evaluating NIRF to assess bone perfusion. Methods: A systematic review, according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline, was performed. Studies up to October 2021 were included. We extracted data regarding the study population, size and design, reported objective fluorescence parameters and the methodology used for fluorescence imaging and processing. Results: Ten articles were included. Studies reported unevenly on the protocol used for NIRF imaging. Five studies reported objective parameters. Absolute and relative perfusion parameters and parameters derived from maximum fluorescence were reported. The clinical significance of these parameters has not been evaluated in humans. Conclusion: The evidence on bone perfusion as measured with NIRF is limited. More clinical studies are required.

Quantitative Fluorescence Imaging of Perfusion-An Algorithm to Predict Anastomotic Leakage.

This study tests fluorescence imaging-derived quantitative parameters for perfusion evaluation of the gastric tube during surgery and correlates these parameters with patient outcomes in terms of anastomotic leakage. Poor fundus perfusion is seen as a major factor for the development of anastomotic leakage and strictures. Fluorescence perfusion imaging may reduce the incidence of complications. Parameters for the quantification of the fluorescence signal are still lacking. Quantitative parameters in terms of maximal intensity, mean slope and influx timepoint were tested for significant differences between four perfusion areas of the gastric tube in 22 patients with a repeated ANOVA test. These parameters were compared with patient outcomes. Maximal intensity, mean slope and influx timepoint were significantly different between the base of the gastric tube and the fundus (p < 0.0001). Patients who developed anastomotic leakage showed a mean slope of almost 0 in Location 4. The distance of the demarcation of ICG to the fundus was significantly higher in the three patients who developed anastomotic leakage (p < 0.0001). This study presents quantitative intra-operative perfusion imaging with fluorescence. Quantification of the fluorescence signal allows for early risk stratification of necrosis.

Deep Learning-based Recurrence Prediction in Patients with Non-muscle-invasive Bladder Cancer.

BACKGROUND: Non-muscle-invasive bladder cancer (NMIBC) is characterized by frequent recurrence of the disease, which is difficult to predict. OBJECTIVE: To combine digital histopathology slides with clinical data to predict 1- and 5-yr recurrence-free survival of NMIBC patients using deep learning. DESIGN, SETTING, AND PARTICIPANTS: Data of patients undergoing a transurethral resection of a bladder tumor between 2000 and 2018 at a Dutch academic medical center were selected. Corresponding histological slides were digitized. A three-step approach was used to predict 1- and 5-yr recurrence-free survival. First, a segmentation network was used to detect the urothelium on the digital histopathology slides. Second, a selection network was trained for the selection of patches associated with recurrence. Third, a classification network, combining the information of the selection network with clinical data, was trained to give the probability of 1- and 5-yr recurrence-free survival. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: The accuracy of the deep learning-based model was compared with a multivariable logistic regression model using clinical data only. RESULTS AND LIMITATIONS: In the 1- and 5-yr follow-up cohorts, 359 and 281 patients were included with recurrence rates of 27% and 63%, respectively. The areas under the curve (AUCs) of the model combining digital histopathology slide data with clinical data were 0.62 and 0.76 for 1- and 5-yr recurrence predictions, respectively, which were higher than those of the model using digital histopathology slide data only (AUCs of 0.56 and 0.72, respectively) and the multivariable logistic regression (AUCs of 0.58 and 0.57, respectively). CONCLUSIONS: In our population, the deep learning-based model combining digital histopathology slides and clinical data enhances the prediction of recurrence (within 5 yr) compared with models using clinical data or image data only. PATIENT SUMMARY: By combining histopathology images and patient record data using deep learning, the prediction of recurrence in bladder cancer patients is enhanced.

Machine learning for grading and prognosis of esophageal dysplasia using mass spectrometry and histological imaging.

BACKGROUND: Barrett's esophagus (BE) is a precursor lesion of esophageal adenocarcinoma and may progress from non-dysplastic through low-grade dysplasia (LGD) to high-grade dysplasia (HGD) and cancer. Grading BE is of crucial prognostic value and is currently based on the subjective evaluation of biopsies. This study aims to investigate the potential of machine learning (ML) using spatially resolved molecular data from mass spectrometry imaging (MSI) and histological data from microscopic hematoxylin and eosin (H&E)-stained imaging for computer-aided diagnosis and prognosis of BE. METHODS: Biopsies from 57 patients were considered, divided into non-dysplastic (n = 15), LGD non-progressive (n = 14), LGD progressive (n = 14), and HGD (n = 14). MSI experiments were conducted at 50 × 50 µm spatial resolution per pixel corresponding to a tile size of 96x96 pixels in the co-registered H&E images, making a total of 144,823 tiles for the whole dataset. RESULTS: ML models were trained to distinguish epithelial tissue from stroma with area-under-the-curve (AUC) values of 0.89 (MSI) and 0.95 (H&E)) and dysplastic grade (AUC of 0.97 (MSI) and 0.85 (H&E)) on a tile level, and low-grade progressors from non-progressors on a patient level (accuracies of 0.72 (MSI) and 0.48 (H&E)). CONCLUSIONS: In summary, while the H&E-based classifier was best at distinguishing tissue types, the MSI-based model was more accurate at distinguishing dysplastic grades and patients at progression risk, which demonstrates the complementarity of both approaches. Data are available via ProteomeXchange with identifier PXD028949.

MRI and CT in the follow-up after irreversible electroporation of small renal masses.

PURPOSE: Ablation plays a growing role in the treatment of small renal masses (SRMs) due to its nephron sparing properties and low invasiveness. Irreversible electroporation (IRE) has the potential, although still experimental, to overcome current limitations of thermal ablation. No prospective imaging studies exist of the ablation zone in the follow up after renal IRE in humans. Objectives are to assess computed tomography (CT) and magnetic resonance imaging (MRI) on the ablation zone volume (AZV), enhancement and imaging characteristics after renal IRE. METHODS: Prospective phase 2 study of IRE in nine patients with ten SRMs. MRI imaging was performed pre-IRE, 1 week, 3 months, 6 months and 12 months after IRE. CT was performed pre-IRE, perioperatively (direct after ablation), 3 months, 6 months and 12 months after IRE. AZVs were assessed by two independent observers. Observer variation was analyzed. Evolution of AZVs, and relation between the needle configuration volume (NCV; planned AZV) and CT- and MRI volumes were evaluated. RESULTS: Eight SRMs were clear cell renal cell carcinomas, one SRM was a papillary renal cell carcinoma and one patient had a non-diagnostic biopsy. On CT, median AZV increased perioperatively until 3 months post-IRE (respectively, 16.8 cm3 and 6.2 cm3) compared to the NCV (4.8 cm3). On MRI, median AZV increased 1-week post-IRE until 3 months post-IRE (respectively, 14.5 cm3 and 4.6 cm3) compared to the NCV (4.8 cm3). At 6 months the AZV starts decreasing (CT 4.8 cm3; MRI 3.0 cm3), continuing at 12 months (CT 4.2 cm3, MRI 1.1 cm3). Strong correlation was demonstrated between the planning and the post-treatment volumes. Inter-observer agreement between observers was excellent (CT 95% CI 0.82-0.95, MRI 95% CI 0.86-0.96). All SRMs appeared non-enhanced immediately after ablation, except for one residual tumour. Subtraction images confirmed non-enhancement on MRI in unclear enhancement cases (3/9). Directly after IRE, gas bubbles, perinephric stranding and edema were observed in all cases. CONCLUSION: The AZV increases immediately on CT until 3 months after IRE. On MRI, the AZV increases at 1 week until 3 months post-IRE. At 6 months the AZV starts decreasing until 12 months post-IRE on both CT and MRI. Enhancement was absent post-IRE, except for one residual tumour. Gas bubbles, perinephric stranding and edema are normal findings directly post-IRE.

Toward improved endoscopic surveillance with multidiameter single fiber reflectance spectroscopy in patients with Barrett's esophagus.

Patients with Barrett's esophagus are at an increased risk to develop esophageal cancer and, therefore, undergo regular endoscopic surveillance. Early detection of neoplasia enables endoscopic treatment, which improves outcomes. However, early Barrett's neoplasia is easily missed during endoscopic surveillance. This study investigates multidiameter single fiber reflectance spectroscopy (MDSFR) to improve Barrett's surveillance. Based on the concept of field cancerization, it may be possible to identify the presence of a neoplastic lesion from measurements elsewhere in the esophagus or even the oral cavity. In this study, MDSFR measurements are performed on non-dysplastic Barrett's mucosa, squamous mucosa, oral mucosa, and the neoplastic lesion (if present). Based on logistic regression analysis on the scattering parameters measured by MDSFR, a classifier is developed that can predict the presence of neoplasia elsewhere in the Barrett's segment from measurements on the non-dysplastic Barrett's mucosa (sensitivity 91%, specificity 71%, AUC = 0.77). Classifiers obtained from logistic regression analysis for the squamous and oral mucosa do not result in an AUC significantly different from 0.5.

Quantitative change of perfusion in gastric tube reconstruction by sidestream dark field microscopy (SDF) after esophagectomy, a prospective in-vivo cohort study.

BACKGROUND: Anastomotic leakage is one of the most severe complications in patients undergoing esophagectomy with gastric tube reconstruction. Transection of the left gastric and gastro-epiploic artery and vein results in compromised perfusion which is seen as the major contributing factor for anastomotic dehiscence. The main objective of this prospective, observational, in-vivo pilot study is to microscopically evaluate gastric tube perfusion with Sidestream Darkfield Microscopy (SDF). METHODS: Intra-operative microscopic images of gastric-microcirculation were obtained with SDF directly after reconstruction in 22 patients. Quantitative perfusion related parameters were: velocity, Microvascular Flow Index(MFI), Total Vessel Density(TVD), Perfusion Vessel Density(PVD), Proportion of Perfused Vessels(PPV) and De Backer Score(DBS). Dedicated software was used to assess parameters predictive for compromised perfusion. RESULTS: SDF was feasible to accurately visualize and evaluate microcirculation in all patients. Velocity(µm/sec) was significantly decreased towards the fundus (p = 0.001). MFI, PVD and PVD were decreased distal of the watershed - between the right and left gastro-epiploic artery and vein - and in the fundus, compared to the base of the gastric tube(p = 0.0002). No differences in TVD and DBS were observed; because of vessel-dilation in the fundus-area. This suggests that venous congestion results in comprised inflow of oxygen rich blood and plays a role in the development of ischaemia. CONCLUSION: We present quantitative perfusion imaging with SDF of the gastric tube. Velocity, MFI, TVD and PPV are accurate parameters to observe perfusion decrease. Also, venous congestion is visible in the fundus, suggesting an important role in the development of ischaemia. These parameters could allow early risk stratification, and, potentially, can accomplish a reduction in anastomotic leakage.

Computed Tomography-Mediated Registration of Trapeziometacarpal Articular Cartilage Using Intraarticular Optical Coherence Tomography and Cryomicrotome Imaging: A Cadaver Study.

OBJECTIVE: Accurate, high-resolution imaging of articular cartilage thickness is an important clinical challenge in patients with osteoarthritis, especially in small joints. In this study, computed tomography (CT) mediated catheter-based optical coherence tomography (OCT) was utilized to create a digital reconstruction of the articular surface of the trapeziometacarpal (TMC) joint and to assess cartilage thickness in comparison to cryomicrotome data. DESIGN: Using needle-based introduction of the OCT probe, the articular surface of the TMC joint of 5 cadaver wrists was scanned in different probe positions with matching CT scans to record the intraarticular probe trajectory. Subsequently and based on the acquired CT data, 3-dimensional realignment of the OCT data to the curved intraarticular trajectory was performed for all probe positions. The scanned TMC joints were processed using a cryomicrotome imaging system. Finally, cartilage thickness measurements between OCT and cryomicrotome data were compared. RESULTS: Successful visualization of TMC articular cartilage was performed using OCT. The CT-mediated registration yielded a digital reconstruction of the articular surface on which thickness measurements could be performed. A near-perfect agreement between OCT and cryomicrotome thickness measurements was found (r2 = 0.989). CONCLUSION: The proposed approach enables 3D reconstruction of the TMC articular surface with subsequent accurate cartilage thickness measurements, encouraging the development of intraarticular cartilage OCT for future (clinical) application.

Automated Detection and Grading of Non-Muscle-Invasive Urothelial Cell Carcinoma of the Bladder.

Accurate grading of non-muscle-invasive urothelial cell carcinoma is of major importance; however, high interobserver variability exists. A fully automated detection and grading network based on deep learning is proposed to enhance reproducibility. A total of 328 transurethral resection specimens from 232 patients were included, and a consensus reading by three specialized pathologists was used. The slides were digitized, and the urothelium was annotated by expert observers. The U-Net-based segmentation network was trained to automatically detect urothelium. This detection was used as input for the classification network. The classification network aimed to grade the tumors according to the World Health Organization grading system adopted in 2004. The automated grading was compared with the consensus and individual grading. The segmentation network resulted in an accurate detection of urothelium. The automated grading shows moderate agreement (κ = 0.48 ± 0.14 SEM) with the consensus reading. The agreement among pathologists ranges between fair (κ = 0.35 ± 0.13 SEM and κ = 0.38 ± 0.11 SEM) and moderate (κ = 0.52 ± 0.13 SEM). The automated classification correctly graded 76% of the low-grade cancers and 71% of the high-grade cancers according to the consensus reading. These results indicate that deep learning can be used for the fully automated detection and grading of urothelial cell carcinoma.

En-face optical coherence tomography for the detection of cancer in prostatectomy specimens: Quantitative analysis in 20 patients.

The increase histopathological evaluation of prostatectomy specimens rises the workload on pathologists. Automated histopathology systems, preferably directly on unstained specimens, would accelerate the pathology workflow. In this study, we investigate the potential of quantitative analysis of optical coherence tomography (OCT) to separate benign from malignant prostate tissue automatically. Twenty fixated prostates were cut, from which 54 slices were scanned by OCT. Quantitative OCT metrics (attenuation coefficient, residue, goodness-of-fit) were compared for different tissue types, annotated on the histology slides. To avoid misclassification, the poor-quality slides, and edges of annotations were excluded. Accurate registration of OCT data with histology was achieved in 31 slices. After removing outliers, 56% of the OCT data was compared with histopathology. The quantitative data could not separate malignant from benign tissue. Logistic regression resulted in malignant detection with a sensitivity of 0.80 and a specificity of 0.34. Quantitative OCT analysis should be improved before clinical use.

Evaluation of acute esophageal radiation-induced damage using magnetic resonance imaging: a feasibility study in mice.

BACKGROUND: Thoracic and head and neck cancer radiation therapy (RT) can cause damage to nearby healthy organs such as the esophagus, causing acute radiation-induced esophageal damage (ARIED). A non-invasive method to detect and monitor ARIED can facilitate optimizing RT to avoid ARIED while improving local tumor control. Current clinical guidelines are limited to scoring the esophageal damage based on the symptoms of patients. Magnetic resonance imaging (MRI) is a non-invasive imaging modality that may potentially visualize radiation-induced organ damage. We investigated the feasibility of using T2-weighted MRI to detect and monitor ARIED using a two-phased study in mice. METHODS: The first phase aimed to establish the optimal dose level at which ARIED is inducible and to determine the time points where ARIED is detectable. Twenty four mice received a single dose delivery of 20 and 40 Gy at proximal and distal spots of 10.0 mm (in diameter) on the esophagus. Mice underwent MRI and histopathology analysis with esophageal resection at two, three, and 4 weeks post-irradiation, or earlier in case mice had to be euthanized due to humane endpoints. In the second phase, 32 mice received a 40 Gy single dose and were studied at two, three, and 7 days post-irradiation. We detected ARIED as a change in signal intensity of the MRI images. We measured the width of the hyperintense area around the esophagus in all mice that underwent MRI prior to and after irradiation. We conducted a blind qualitative comparison between MRI findings and histopathology as the gold standard. RESULTS/CONCLUSIONS: A dose of 40 Gy was needed to induce substantial ARIED. MRI detected ARIED as high signal intensity, visible from 2 days post-irradiation. Quantitative MRI analysis showed that the hyperintense area around the esophagus with severe ARIED was 1.41 mm wider than with no damage and MRI-only mice. The overall sensitivity and specificity were 56 and 43% respectively to detect any form of ARIED. However, in this study MRI correctly detected 100% of severe ARIED cases. Our two-phased preclinical study showed that MRI has the potential to detect ARIED as a change in signal intensity and width of enhancement around the esophagus.

Toward Automated In Vivo Bladder Tumor Stratification Using Confocal Laser Endomicroscopy.

Purpose: Urothelial carcinoma of the bladder (UCB) is the most common urinary cancer. White-light cystoscopy (WLC) forms the corner stone for the diagnosis of UCB. However, histopathological assessment is required for adjuvant treatment selection. Probe-based confocal laser endomicroscopy (pCLE) enables visualization of the microarchitecture of bladder lesions during WLC, which allows for real-time tissue differentiation and grading of UCB. To improve the diagnostic process of UCB, computer-aided classification of pCLE videos of in vivo bladder lesions were evaluated in this study. Materials and Methods: We implemented preprocessing methods to optimize contrast and to reduce striping artifacts in each individual pCLE frame. Subsequently, a semiautomatic frame selection was performed. The selected frames were used to train a feature extractor based on pretrained ImageNet networks. A recurrent neural network, in specific long short-term memory (LSTM), was used to predict the grade of bladder lesions. Differentiation of lesions was performed at two levels, namely (i) healthy and benign vs malignant tissue and (ii) low-grade vs high-grade papillary UCB. A total of 53 patients with 72 lesions were included in this study, resulting in ∼140,000 pCLE frames. Results: The semiautomated frame selection reduced the number of frames to ∼66,500 informative frames. The accuracy for differentiation of (i) healthy and benign vs malignant urothelium was 79% and (ii) high-grade and low-grade papillary UCB was 82%. Conclusions: A feature extractor in combination with LSTM results in proper stratification of pCLE videos of in vivo bladder lesions.