There is a recent and increasing trend in the incidence of pancreatic neuroendocrine tumors (PNETs). Ki-67 proliferative index is required for routine pathologic evaluation of PNETs. This index has been found to be a consistent prognostic factor to assess clinical/prognostic outcome of PNETs. Unfortunately, we still lack a standardized method to reliably obtain the Ki-67 proliferative index. As part of a large study to standardize this index, here we present an accurate, easy-to- use, reproducible method to identify tumor nuclei and hotspots within PNETs. We modified the U-Net image segmentation architecture to identify tumor positive and negative nuclei. We also introduced the concept of local depth for identification of hotspots. On an independent test set of 8 whole slide images, the modified U-Net achieved a sensitivity of 96.2% and specificity of 93.3%. The hotspot detection framework resulted in a dice coefficient of 0.81. The method has the potential to not only facilitate the detection of tumor nuclei, but can be adapted to reproduce hotspots by pathologists.
The morphological features that pathologists use to differentiate neoplasms from normal tissue are nonspecific to tissue type. For example, if given a Ki67 stained biopsy of neuroendocrine or breast tumor, a pathologist would be able to correctly identify morphologically abnormal cells in both samples but may struggle to identify the origin of both samples. This is also true for other pathological malignancies such as carcinomas, sarcomas, and leukemia. This implies that computer algorithms trained to recognize tumor from one site should be able to identify tumor from other sites with similar tumor subtypes. Here, we present the results of an experiment that supports this hypothesis. We train a deep learning system to distinguish tumor from non-tumor regions in Ki67 stained neuroendocrine tumor digital slides. Then, we test the same, unmodified, deep learning model to distinguish breast cancer from non-cancer regions. When applied to a sample of 96 high power fields, our system achieved a cumulative pixel-wise accuracy of 86% across these high-power fields. To our knowledge, our results are the first to formally demonstrate generalized segmentation of tumors from different sites of origin through image analysis. This paradigm has the potential to help with the design of tumor identification algorithms as well as the composition of the datasets they draw from.
Pathology remains the gold standard for cancer diagnosis and in some cases prognosis, in which trained pathologists examine abnormality in tissue architecture and cell morphology characteristic of cancer cells with a bright-field microscope. The limited resolution of conventional microscope can result in intra-observer variation, missed early-stage cancers, and indeterminate cases that often result in unnecessary invasive procedures in the absence of cancer. Assessment of nanoscale structural characteristics via quantitative phase represents a promising strategy for identifying pre-cancerous or cancerous cells, due to its nanoscale sensitivity to optical path length, simple sample preparation (i.e., label-free) and low cost. I will present the development of quantitative phase microscopy system in transmission and reflection configuration to detect the structural changes in nuclear architecture, not be easily identifiable by conventional pathology. Specifically, we will present the use of transmission-mode quantitative phase imaging to improve diagnostic accuracy of urine cytology and the nuclear dry mass is progressively correlate with negative, atypical, suspicious and positive cytological diagnosis. In a second application, we will present the use of reflection-mode quantitative phase microscopy for depth-resolved nanoscale nuclear architecture mapping (nanoNAM) of clinically prepared formalin-fixed, paraffin-embedded tissue sections. We demonstrated that the quantitative phase microscopy system detects a gradual increase in the density alteration of nuclear architecture during malignant transformation in animal models of colon carcinogenesis and in human patients with ulcerative colitis, even in tissue that appears histologically normal according to pathologists. We evaluated the ability of nanoNAM to predict "future" cancer progression in patients with ulcerative colitis.
The grading of neuroendocrine tumors of the digestive system is dependent on accurate and reproducible assessment of the proliferation with the tumor, either by counting mitotic figures or counting Ki-67 positive nuclei. At the moment, most pathologists manually identify the hotspots, a practice which is tedious and irreproducible. To better help pathologists, we present an automatic method to detect all potential hotspots in neuroendocrine tumors of the digestive system. The method starts by segmenting Ki-67 positive nuclei by entropy based thresholding, followed by detection of centroids for all Ki-67 positive nuclei. Based on geodesic distance, approximated by the nuclei centroids, we compute two maps: an amoeba map and a weighted amoeba map. These maps are later combined to generate the heat map, the segmentation of which results in the hotspots. The method was trained on three and tested on nine whole slide images of neuroendocrine tumors. When evaluated by two expert pathologists, the method reached an accuracy of 92.6%. The current method does not discriminate between tumor, stromal and inflammatory nuclei. The results show that α-shape maps may represent how hotspots are perceived.
Effective management of patients who are at risk of developing invasive cancer is a primary challenge in early cancer
detection. Techniques that can help establish clear-cut protocols for successful triaging of at-risk patients have the
potential of providing critical help in improving patient care while simultaneously reducing patient cost. We have
developed such a technique for early prediction of cancer progression that uses unstained tissue sections to provide
depth-resolved nanoscale nuclear architecture mapping (nanoNAM) of heterogeneity in optical density alterations
manifested in precancerous lesions during cancer progression. We present nanoNAM and its application to predicting
cancer progression in a well-established mouse model of spontaneous carcinogenesis: ApcMin/+ mice.
The development of accurate and clinically applicable tools to assess cancer risk is essential to define candidates to undergo screening for early-stage cancers at a curable stage or provide a novel method to monitor chemoprevention treatments. With the use of our recently developed optical technology—spatial-domain low-coherence quantitative phase microscopy (SL-QPM), we have derived a novel optical biomarker characterized by structure-derived optical path length (OPL) properties from the cell nucleus on the standard histology and cytology specimens, which quantifies the nano-structural alterations within the cell nucleus at the nanoscale sensitivity, referred to as nano-morphology marker. The aim of this study is to evaluate the feasibility of the nuclear nano-morphology marker from histologically normal cells, extracted directly from the standard histology specimens, to detect early-stage carcinogenesis, assess cancer risk, and monitor the effect of chemopreventive treatment. We used a well-established mouse model of spontaneous carcinogenesis—ApcMin mice, which develop multiple intestinal adenomas (Min) due to a germline mutation in the adenomatous polyposis coli (Apc) gene. We found that the nuclear nano-morphology marker quantified by OPL detects the development of carcinogenesis from histologically normal intestinal epithelial cells, even at an early pre-adenomatous stage (six weeks). It also exhibits a good temporal correlation with the small intestine that parallels the development of carcinogenesis and cancer risk. To further assess its ability to monitor the efficacy of chemopreventive agents, we used an established chemopreventive agent, sulindac. The nuclear nano-morphology marker is reversed toward normal after a prolonged treatment. Therefore, our proof-of-concept study establishes the feasibility of the SL-QPM derived nuclear nano-morphology marker OPL as a promising, simple and clinically applicable biomarker for cancer risk assessment and evaluation of chemopreventive treatment.
For any technique to be adopted into a clinical setting, it is imperative that it seamlessly integrates with well-established clinical diagnostic workflow. We recently developed an optical microscopy technique-spatial-domain low-coherence quantitative phase microscopy (SL-QPM) that can extract the refractive index of the cell nucleus from the standard histology specimens on glass slides prepared via standard clinical protocols. This technique has shown great potential in detecting cancer with a better sensitivity than conventional pathology. A major hurdle in the clinical translation of this technique is the intrinsic variation among staining agents used in histology specimens, which limits the accuracy of refractive index measurements of clinical samples. In this paper, we present a simple and easily generalizable method to remove the effect of variations in staining levels on nuclear refractive index obtained with SL-QPM. We illustrate the efficacy of our correction method by applying it to variously stained histology samples from animal model and clinical specimens.
Intrigued by our recent finding that the nuclear refractive index is significantly increased in malignant cells and histologically normal cells in clinical histology specimens derived from cancer patients, we sought to identify potential biological mechanisms underlying the observed phenomena. The cell cycle is an ordered series of events that describes the intervals of cell growth, DNA replication, and mitosis that precede cell division. Since abnormal cell cycles and increased proliferation are characteristic of many human cancer cells, we hypothesized that the observed increase in nuclear refractive index could be related to an abundance or accumulation of cells derived from cancer patients at a specific point or phase(s) of the cell cycle. Here we show that changes in nuclear refractive index of fixed cells are seen as synchronized populations of cells that proceed through the cell cycle, and that increased nuclear refractive index is strongly correlated with increased DNA content. We therefore propose that an abundance of cells undergoing DNA replication and mitosis may explain the increase in nuclear refractive index observed in both malignant and histologically normal cells from cancer patients. Our findings suggest that nuclear refractive index may be a novel physical parameter for early cancer detection and risk stratification.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.