Use of topical methylene blue to image nuclear morphometry with a low-cost scanning darkfield microendoscope.

Significance: Fiber-optic microendoscopy is a promising approach to noninvasively visualize epithelial nuclear morphometry for early cancer and precancer detection. However, the broader clinical application of this approach is limited by a lack of topical contrast agents available for in vivo use. Aim: The aim of this study was to evaluate the ability to image nuclear morphometry in vivo with a novel fiber-optic microendoscope used together with topical application of methylene blue (MB), a dye with FDA approval for use in chromoendoscopy in the gastrointestinal tract. Approach: The low-cost, high-resolution microendoscope implements scanning darkfield imaging without complex optomechanical components by leveraging programmable illumination and the rolling shutter of the image sensor. We validate the integration of our system and MB staining for visualizing epithelial cell nuclei by performing ex vivo imaging on fresh animal specimens and in vivo imaging on healthy volunteers. Results: The results indicate that scanning darkfield imaging significantly reduces specular reflection and resolves epithelial nuclei with enhanced image contrast and spatial resolution compared to non-scanning widefield imaging. The image quality of darkfield images with MB staining is comparable to that of fluorescence images with proflavine staining. Conclusions: Our approach enables real-time microscopic evaluation of nuclear patterns and has the potential to be a powerful noninvasive tool for early cancer detection.

DeepDOF-SE: affordable deep-learning microscopy platform for slide-free histology.

Histopathology plays a critical role in the diagnosis and surgical management of cancer. However, access to histopathology services, especially frozen section pathology during surgery, is limited in resource-constrained settings because preparing slides from resected tissue is time-consuming, labor-intensive, and requires expensive infrastructure. Here, we report a deep-learning-enabled microscope, named DeepDOF-SE, to rapidly scan intact tissue at cellular resolution without the need for physical sectioning. Three key features jointly make DeepDOF-SE practical. First, tissue specimens are stained directly with inexpensive vital fluorescent dyes and optically sectioned with ultra-violet excitation that localizes fluorescent emission to a thin surface layer. Second, a deep-learning algorithm extends the depth-of-field, allowing rapid acquisition of in-focus images from large areas of tissue even when the tissue surface is highly irregular. Finally, a semi-supervised generative adversarial network virtually stains DeepDOF-SE fluorescence images with hematoxylin-and-eosin appearance, facilitating image interpretation by pathologists without significant additional training. We developed the DeepDOF-SE platform using a data-driven approach and validated its performance by imaging surgical resections of suspected oral tumors. Our results show that DeepDOF-SE provides histological information of diagnostic importance, offering a rapid and affordable slide-free histology platform for intraoperative tumor margin assessment and in low-resource settings.

The Feasibility and Outcome of Integra® Bilayer Matrix in the Reconstruction of Oral Cavity Defects.

OBJECTIVE: To evaluate the feasibility, safety, and failure rate of Integra® Bilayer Wound Matrix (Integra) in the reconstruction of oral cavity defects. STUDY DESIGN: Retrospective cohort study. SETTING: All study information was collected from a single academic tertiary care hospital. METHODS: Subjects included adult patients who underwent oral cavity resection and immediate subsequent reconstruction with Integra® Bilayer Wound Matrix at MD Anderson Cancer Center between the years 2015 and 2020. The following variables were collected: patient's demographics, comorbidities, disease stage, treatment and reconstruction modalities, and surgical outcome from the medical records. Statistical analysis included distribution analysis for all collected parameters and Pearson's χ2 tests to find correlation between variables and take rate of Integra. RESULTS: Eighty-three patients underwent reconstruction with Integra® Bilayer Wound Matrix dressing. Average age was 66 years old. Thirty-nine patients (47%) had history of previous resections for oral cavity tumors. Fourteen patients (17%) had history of radiation therapy to the Head and Neck region. Most common pathology was invasive squamous cell carcinoma (75%) followed by dysplasia (12%). Complete wound healing with good cellular integration occurred in 83 patients (96%) with only 3 failures requiring additional surgery. Reconstruction of mandibulectomy defects was associated with increased risk of dehiscence and bone exposure (0.66, P = .03). CONCLUSION: This study shows promising results with high take rate of Integra® Bilayer Wound Matrix dressing in the reconstruction of various oral cavity defects. We encourage surgeons to adopt this technique as a viable and versatile option into the reconstruction ladder of oral cavity defects.

Scanning darkfield high-resolution microendoscope for label-free microvascular imaging.

Characterization of microvascular changes during neoplastic progression has the potential to assist in discriminating precancer and early cancer from benign lesions. Here, we introduce a novel high-resolution microendoscope that leverages scanning darkfield reflectance imaging to characterize angiogenesis without exogenous contrast agents. Scanning darkfield imaging is achieved by coupling programmable illumination with a complementary metal-oxide semiconductor (CMOS) camera rolling shutter, eliminating the need for complex optomechanical components and making the system portable, low-cost (<$5,500) and simple to use. Imaging depth is extended by placing a gradient-index (GRIN) lens at the distal end of the imaging fiber to resolve subepithelial microvasculature. We validated the capability of the scanning darkfield microendoscope to visualize microvasculature at different anatomic sites in vivo by imaging the oral cavity of healthy volunteers. Images of cervical specimens resected for suspected neoplasia reveal distinct microvascular patterns in columnar and squamous epithelium with different grades of precancer, indicating the potential of scanning darkfield microendoscopy to aid in efforts to prevent cervical cancer through early diagnosis.

A Deep Learning Onion Peeling Approach to Measure Oral Epithelium Layer Number.

Head and neck squamous cell carcinoma (HNSCC), specifically in the oral cavity (oral squamous cell carcinoma, OSCC), is a common, complex cancer that significantly affects patients' quality of life. Early diagnosis typically improves prognoses yet relies on pathologist examination of histology images that exhibit high inter- and intra-observer variation. The advent of deep learning has automated this analysis, notably with object segmentation. However, techniques for automated oral dysplasia diagnosis have been limited to shape or cell stain information, without addressing the diagnostic potential in counting the number of cell layers in the oral epithelium. Our study attempts to address this gap by combining the existing U-Net and HD-Staining architectures for segmenting the oral epithelium and introducing a novel algorithm that we call Onion Peeling for counting the epithelium layer number. Experimental results show a close correlation between our algorithmic and expert manual layer counts, demonstrating the feasibility of automated layer counting. We also show the clinical relevance of oral epithelial layer number to grading oral dysplasia severity through survival analysis. Overall, our study shows that automated counting of oral epithelium layers can represent a potential addition to the digital pathology toolbox. Model generalizability and accuracy could be improved further with a larger training dataset.

Deep learning-based pathology image analysis predicts cancer progression risk in patients with oral leukoplakia.

BACKGROUND: Oral leukoplakia (OL) is associated with an increased risk for oral cancer (OC) development. Prediction of OL cancer progression may contribute to decreased OC morbidity and mortality by favoring early intervention. Current OL progression risk assessment approaches face large interobserver variability and is weakly prognostic. We hypothesized that convolutional neural networks (CNN)-based histology image analyses could accelerate the discovery of better OC progression risk models. METHODS: Our CNN-based oral mucosa risk stratification model (OMRS) was trained to classify a set of nondysplastic oral mucosa (OM) and a set of OC H&E slides. As a result, the OMRS model could identify abnormal morphological features of the oral epithelium. By applying this model to OL slides, we hypothesized that the extent of OC-like features identified in the OL epithelium would correlate with its progression risk. The OMRS model scored and categorized the OL cohort (n = 62) into high- and low-risk groups. RESULTS: OL patients classified as high-risk (n = 31) were 3.98 (95% CI 1.36-11.7) times more likely to develop OC than low-risk ones (n = 31). Time-to-progression significantly differed between high- and low-risk groups (p = 0.003). The 5-year OC development probability was 21.3% for low-risk and 52.5% for high-risk patients. The predictive power of the OMRS model was sustained even after adjustment for age, OL site, and OL dysplasia grading (HR = 4.52, 1.5-13.7). CONCLUSION: The ORMS model successfully identified OL patients with a high risk of OC development and can potentially benefit OC early diagnosis and prevention policies.

Spatial PD-L1, immune-cell microenvironment, and genomic copy-number alteration patterns and drivers of invasive-disease transition in prospective oral precancer cohort.

BACKGROUND: Studies of the immune landscape led to breakthrough trials of programmed death-1 (PD-1) inhibitors for recurrent/metastatic head and neck squamous cell carcinoma therapy. This study investigated the timing, influence of somatic copy-number alterations (SCNAs), and clinical implications of PD-L1 and immune-cell patterns in oral precancer (OPC). METHODS: The authors evaluated spatial CD3, CD3/8, and CD68 density (cells/mm2 ) and PD-L1 (membranous expression in cytokeratin-positive intraepithelial neoplastic cells and CD68) patterns by multiplex immunofluorescence in a 188-patient prospective OPC cohort, characterized by clinical, histologic, and SCNA risk factors and protocol-specified primary end point of invasive cancer. The authors used Wilcoxon rank-sum and Fisher exact tests, linear mixed effect models, mediation, and Cox regression and recursive-partitioning analyses. RESULTS: Epithelial, but not CD68 immune-cell, PD-L1 expression was detected in 28% of OPCs, correlated with immune-cell infiltration, 9p21.3 loss of heterozygosity (LOH), and inferior oral cancer-free survival (OCFS), notably in OPCs with low CD3/8 cell density, dysplasia, and/or 9p21.3 LOH. High CD3/8 cell density in dysplastic lesions predicted better OCFS and eliminated the excess risk associated with prior oral cancer and dysplasia. PD-L1 and CD3/8 patterns revealed inferior OCFS in PD-L1 high intrinsic induction and dysplastic immune-cold subgroups. CONCLUSION: This report provides spatial insight into the immune landscape and drivers of OPCs, and a publicly available immunogenomic data set for future precancer interrogation. The data suggest that 9p21.3 LOH triggers an immune-hot inflammatory phenotype; whereas increased 9p deletion size encompassing CD274 at 9p24.1 may contribute to CD3/8 and PD-L1 depletion during invasive transition. The inferior OCFS in PD-L1-high, immune-cold OPCs support the development of T-cell recruitment strategies.

Multimodal optical imaging with real-time projection of cancer risk and biopsy guidance maps for early oral cancer diagnosis and treatment.

Significance: Despite recent advances in multimodal optical imaging, oral imaging systems often do not provide real-time actionable guidance to the clinician who is making biopsy and treatment decisions. Aim: We demonstrate a low-cost, portable active biopsy guidance system (ABGS) that uses multimodal optical imaging with deep learning to directly project cancer risk and biopsy guidance maps onto oral mucosa in real time. Approach: Cancer risk maps are generated based on widefield autofluorescence images and projected onto the at-risk tissue using a digital light projector. Microendoscopy images are obtained from at-risk areas, and multimodal image data are used to calculate a biopsy guidance map, which is projected onto tissue. Results: Representative patient examples highlight clinically actionable visualizations provided in real time during an imaging procedure. Results show multimodal imaging with cancer risk and biopsy guidance map projection offers a versatile, quantitative, and precise tool to guide biopsy site selection and improve early detection of oral cancers. Conclusions: The ABGS provides direct visible guidance to identify early lesions and locate appropriate sites to biopsy within those lesions. This represents an opportunity to translate multimodal imaging into real-time clinically actionable visualizations to help improve patient outcomes.

Mildly dysplastic oral lesions with optically-detectable abnormalities share genetic similarities with severely dysplastic lesions.

OBJECTIVE: Optical imaging studies of oral premalignant lesions have shown that optical markers, including loss of autofluorescence and altered morphology of epithelial cell nuclei, are predictive of high-grade pathology. While these optical markers are consistently positive in lesions with moderate/severe dysplasia or cancer, they are positive only in a subset of lesions with mild dysplasia. This study compared the gene expression profiles of lesions with mild dysplasia (stratified by optical marker status) to lesions with severe dysplasia and without dysplasia. MATERIALS AND METHODS: Forty oral lesions imaged in patients undergoing oral surgery were analyzed: nine without dysplasia, nine with severe dysplasia, and 22 with mild dysplasia. Samples were submitted for high throughput gene expression analysis. RESULTS: The analysis revealed 116 genes differentially expressed among sites without dysplasia and sites with severe dysplasia; 50 were correlated with an optical marker quantifying altered nuclear morphology. Ten of 11 sites with mild dysplasia and positive optical markers (91%) had gene expression similar to sites with severe dysplasia. Nine of 11 sites with mild dysplasia and negative optical markers (82%) had similar gene expression as sites without dysplasia. CONCLUSION: This study suggests that optical imaging may help identify patients with mild dysplasia who require more intensive clinical follow-up. If validated, this would represent a significant advance in patient care for patients with oral premalignant lesions.

Outcomes for recurrent oral cavity squamous cell carcinoma.

OBJECTIVES: We sought to determine overall survival (OS), prognostic factors, cost, and functional outcomes after surgery for locally recurrent oral cavity squamous cell carcinoma (OCSCC). MATERIALS AND METHODS: We retrospectively reviewed 399 cases of locally recurrent OCSCC from 1997 to 2011, of which 259 patients were treated with salvage surgery. Survival and prognostic factors were evaluated using univariable and multivariable Cox regression, the Kaplan-Meier method, and the log-rank test. RESULTS: The 5-year OS for patients undergoing surgical salvage, nonsurgical therapy, or supportive care was 44.2%, 1.5%, and 0%, respectively. For patients who underwent surgical salvage, 133 (51%) patients experienced a second recurrence at a median of 17 months. Factors associated with OS included disease-free interval ≤ 6 months (P =.0001), recurrent stage III-IV disease (P <.0001), and prior radiation (P =.0001). Patients with both advanced stage and prior radiation had a 23% 5-year OS, compared with 70% for those with neither risk (P <.001). Functionally, 85% of patients had > 80% speech intelligibility and 81% were able to eat by mouth following salvage surgery. Of the patients who required tracheostomy, 78% were decannulated. The adjusted median hospital and professional charges for patients were $129,696 (range $9,238-$956,818). CONCLUSIONS: Patients with recurrent OCSCC who underwent salvage surgery have favorable functional outcomes with half of alive at 5 years but poorer OS for advanced disease, disease-free interval ≤ 6 months, and prior radiation. Additionally, treatment is associated with high cost, and about half of patients ultimately develop another recurrence.

Outcomes after definitive surgery for mandibular osteoradionecrosis.

OBJECTIVES: To analyze charges, complications, survival, and functional outcomes for definitive surgery of mandibular osteoradionecrosis (ORN). MATERIALS AND METHODS: Retrospective analysis of 76 patients who underwent segmental mandibulectomy with reconstruction from 2000 to 2009. RESULTS: Complications occurred in 49 (65%) patients and were associated with preoperative drainage (odds ratio [OR] 4.40, 95% confidence interval [CI] 1.01-19.27). The adjusted median charge was $343 000, and higher charges were associated with double flap reconstruction (OR 8.15, 95% CI 2.19-30.29) and smoking (OR 5.91, 95% CI 1.69-20.72). Improved swallow was associated with age <67 years (OR 3.76, 95% CI 1.16-12.17) and preoperative swallow (OR 3.42, 95% CI 1.23-9.51). Five-year ORN-recurrence-free survival was 93% while overall survival was 63% and associated with pulmonary disease (HR [hazard ratio] 3.57, 95% CI 1.43-8.94). CONCLUSIONS: Although recurrence of ORN is rare, surgical complications are common and charges are high. Poorer outcomes and higher charges are associated with preoperative factors.

Phase II Clinical Trial of Neoadjuvant and Adjuvant Pembrolizumab in Resectable Local-Regionally Advanced Head and Neck Squamous Cell Carcinoma.

PURPOSE: Patients with resected, local-regionally advanced, head and neck squamous cell carcinoma (HNSCC) have a one-year disease-free survival (DFS) rate of 65%-69% despite adjuvant (chemo)radiotherapy. Neoadjuvant PD-1 immune-checkpoint blockade (ICB) has demonstrated clinical activity, but biomarkers of response and effect on survival remain unclear. PATIENTS AND METHODS: Eligible patients had resectable squamous cell carcinoma of the oral cavity, larynx, hypopharynx, or oropharynx (p16-negative) and clinical stage T3-T4 and/or two or more nodal metastases or clinical extracapsular nodal extension (ENE). Patients received neoadjuvant pembrolizumab 200 mg 1-3 weeks prior to surgery, were stratified by absence (intermediate-risk) or presence (high-risk) of positive margins and/or ENE, and received adjuvant radiotherapy (60-66 Gy) and concurrent pembrolizumab (every 3 weeks × 6 doses). Patients with high-risk HNSCC also received weekly, concurrent cisplatin (40 mg/m2). Primary outcome was one-year DFS. Secondary endpoints were one-year overall survival (OS) and pathologic response (PR). Safety was evaluated with CTCAE v5.0. RESULTS: From February 2016 to October 2020, 92 patients enrolled. The median age was 59 years (range, 27-80), 30% were female, 86% had stage T3-T4, and 69% had ≥N2. At a median follow-up of 28 months, one-year DFS was 97% (95% CI, 71%-90%) in the intermediate-risk group and 66% (95% CI, 55%-84%) in the high-risk group. Patients with a PR had significantly improved one-year DFS relative to patients without response (93% vs. 72%, hazard ratio 0.29; 95% CI, 11%-77%). No new safety signals were identified. CONCLUSIONS: Neoadjuvant and adjuvant pembrolizumab increased one-year DFS rate in intermediate-risk, but not high-risk, HNSCC relative to historical control. PR to neoadjuvant ICB is a promising surrogate for DFS.

Phase 1/2a, open-label, multicenter study of RM-1929 photoimmunotherapy in patients with locoregional, recurrent head and neck squamous cell carcinoma.

BACKGROUND: Recurrent head and neck squamous cell carcinoma (rHNSCC) represents a significant global health burden with an unmet medical need. In this study we determined the safety and efficacy of RM-1929 photoimmunotherapy in patients with heavily pretreated rHNSCC. METHODS: RM-1929 (anti-EGFR-IR700 dye conjugate) was infused, followed by tumor illumination. We evaluated safety, tumor response, and pharmacokinetics. RESULTS: Nine patients were enrolled in Part 1 (dose-finding) and 30 patients in Part 2 (safety and efficacy). No dose-limiting toxicities were experienced in Part 1; 640 mg/m2 with fixed light dose (50 J/cm2 or 100 J/cm) was recommended for Part 2. Adverse events (AEs) in Part 2 were mostly mild to moderate but 19 (63.3%) patients had AE ≥Grade 3, including 3 (10.0%) with serious AEs leading to death (not treatment related). Efficacy in Part 2: unconfirmed objective response rate (ORR) 43.3% (95% CI 25.46%-62.57%); confirmed ORR 26.7% (95% CI 12.28%-45.89%); median overall survival 9.30 months (95% CI 5.16-16.92 months). CONCLUSIONS: Treatment was well tolerated. Responses and survival following RM-1929 photoimmunotherapy in heavily pretreated patients with rHNSCC were clinically meaningful and warrant further investigation. CLINICAL TRIAL INFORMATION: NCT02422979.

Screening for HPV-related oropharyngeal, anal, and penile cancers in middle-aged men: Initial report from the HOUSTON clinical trial.

BACKGROUND: The Human papillomavirus (HPV)-related Oropharyngeal and Uncommon Cancers Screening Trial of Men (HOUSTON) was designed to determine the prevalence of IgG antibodies to HPV type 16 E proteins (HPV16EAbs), to screen for persistence of HPV and/or detect HPV-related premalignancies and cancers, and to assess acceptance of screening among middle-aged men. METHODS: HOUSTON consists of a cross-sectional study and a longitudinal cohort study of men aged 50-64 years. Serologic HPV16EAb status and oral rinse HPV16 status were determined. All HPV16EAb-positive (HPV16EAb+) men and a matched cohort of HPV16EAb-negative (HPV16EAb-) men as well as all oral rinse HPV16-positive (HPV16+) men were included in the longitudinal study (blinded to their results) and underwent oropharyngeal screening every 6 months as well as one-time anal and penile screening. RESULTS: Of 553 men enrolled in the cross-sectional study, six (1.1%) were HPV16EAb+ (two were also oral rinse HPV16+), and 41 (7.4%) were HPV16EAb- but oral rinse HPV16+. These 47 men, along with five matched controls, were invited to participate in the longitudinal study, and 42 (81%) agreed and completed baseline in-person screening, with 93% and 90% completeing 6-month and 12-month follow-up visits. One HPV16EAb+ (also oral rinse HPV16+) man, who declined participation in the longitudinal study, presented 4 months after enrollment with an early-stage HPV16-related pharyngeal cancer. Additionally, one HPV16EAb+ (oral rinse HPV16-) man and two oral rinse HPV16+ (HPV16EAb-) men were diagnosed with oncogenic HPV-associated anal dysplasia. CONCLUSIONS: This biomarker panel deserves further prospective study to explore potential utility for HPV-related cancer screening among men.

Role of induction chemotherapy for oral cavity squamous cell carcinoma.

BACKGROUND: Patients with locoregionally advanced oral cavity squamous cell carcinoma (OCSCC) have a poor survival outcome. Treatment involves extensive surgery, adjuvant radiation, or chemoradiation and results in high morbidity. In this study, the authors' objective was to evaluate their experience with induction chemotherapy (IC) in the treatment of locoregionally advanced OCSCC. METHODS: A retrospective review of the medical records of all patients with locoregionally advanced (stage III and IV) OCSCC who received IC followed by definitive local therapy was conducted. Outcomes included response to IC and survival. RESULTS: In total, 120 patients were included in the study. The overall stage was stage IV in 79.2% of patients. After 2 cycles of IC, 76 patients (63.3%) achieved at least a partial response, including 13 who had a complete response. Stable disease was observed in 30 patients (25%), and 14 patients (11.7%) had progressive disease. Among responders, 16 patients received definitive chemoradiation or radiation therapy, and 60 underwent surgical resection, of whom 15 had less extensive surgery than was originally planned. Overall, organ preservation was achieved in 40.8% of patients who had a favorable response to IC. The 5-year overall and disease-specific survival rates were 51.4% and 66.9%, respectively. Patients who had at least a partial response had better 5-year overall survival (60.1%) and disease-specific survival (78.5%) compared with nonresponders (33.8% and 46.4%, respectively). CONCLUSIONS: The results demonstrate a response rate to IC in patients with advanced OCSCC similar to what has been observed in patients with cancer in other head and neck subsites. Patients who achieved at least a partial response to IC had a more favorable outcome, with ensuing organ preservation. Further studies are warranted.

Tertiary lymphoid structures with overlapping histopathologic features of cutaneous marginal zone lymphoma during neoadjuvant cemiplimab therapy are associated with antitumor response.

The development of immune checkpoint inhibitor (ICI) therapy with anti-CTLA-4 and anti-PD-1/L1 monoclonal antibodies has led to a paradigm shift in cancer therapy. ICI neoadjuvant therapy followed by surgery has become the standard of care for several advanced-stage cancers. The pathology associated with ICI therapy is vast and includes neoadjuvant-associated tissue reactions and activation of tertiary lymphoid structures (TLSs) at the site of the tumor bed and off-target immune-related adverse events. TLSs are thought to recapitulate lymph node function and may act as localized immune machinery to mount an antitumor response. B-cell activation in TLSs during neoadjuvant ICI therapy has been correlated with antitumor response. We report a patient with a history of sarcomatoid squamous cell carcinoma treated with neoadjuvant ICI cemiplimab who developed clonal expansion of B-cells in the TLSs of the tumor bed. The TLSs morphologically mimicked a cutaneous marginal zone lymphoma with plasmacytic differentiation. Awareness of clonal expansion of B-cells in TLSs during neoadjuvant ICI therapy is critical to recognize a response to ICI therapy and to avoiding an incorrect diagnosis of low-grade B-cell lymphoma.

Deep learning extended depth-of-field microscope for fast and slide-free histology.

Microscopic evaluation of resected tissue plays a central role in the surgical management of cancer. Because optical microscopes have a limited depth-of-field (DOF), resected tissue is either frozen or preserved with chemical fixatives, sliced into thin sections placed on microscope slides, stained, and imaged to determine whether surgical margins are free of tumor cells-a costly and time- and labor-intensive procedure. Here, we introduce a deep-learning extended DOF (DeepDOF) microscope to quickly image large areas of freshly resected tissue to provide histologic-quality images of surgical margins without physical sectioning. The DeepDOF microscope consists of a conventional fluorescence microscope with the simple addition of an inexpensive (less than $10) phase mask inserted in the pupil plane to encode the light field and enhance the depth-invariance of the point-spread function. When used with a jointly optimized image-reconstruction algorithm, diffraction-limited optical performance to resolve subcellular features can be maintained while significantly extending the DOF (200 µm). Data from resected oral surgical specimens show that the DeepDOF microscope can consistently visualize nuclear morphology and other important diagnostic features across highly irregular resected tissue surfaces without serial refocusing. With the capability to quickly scan intact samples with subcellular detail, the DeepDOF microscope can improve tissue sampling during intraoperative tumor-margin assessment, while offering an affordable tool to provide histological information from resected tissue specimens in resource-limited settings.

Algorithm to quantify nuclear features and confidence intervals for classification of oral neoplasia from high-resolution optical images.

Purpose: In vivo optical imaging technologies like high-resolution microendoscopy (HRME) can image nuclei of the oral epithelium. In principle, automated algorithms can then calculate nuclear features to distinguish neoplastic from benign tissue. However, images frequently contain regions without visible nuclei, due to biological and technical factors, decreasing the data available to and accuracy of image analysis algorithms. Approach: We developed the nuclear density-confidence interval (ND-CI) algorithm to determine if an HRME image contains sufficient nuclei for classification, or if a better image is required. The algorithm uses a convolutional neural network to exclude image regions without visible nuclei. Then the remaining regions are used to estimate a confidence interval (CI) for the number of abnormal nuclei per mm 2 , a feature used by a previously developed algorithm (called the ND algorithm), to classify images as benign or neoplastic. The range of the CI determines whether the ND-CI algorithm can classify an image with confidence, and if so, the predicted category. The ND and ND-CI algorithm were compared by calculating their positive predictive value (PPV) and negative predictive value (NPV) on 82 oral biopsies with histopathologically confirmed diagnoses. Results: After excluding the images that could not be classified with confidence, the ND-CI algorithm had higher PPV (65% versus 59%) and NPV (78% versus 75%) than the ND algorithm. Conclusions: The ND-CI algorithm could improve the real-time classification of HRME images of the oral epithelium by informing the user if an improved image is required for diagnosis.