Bladder cancer (BC) is frequent cancer affecting the urinary tract and is one of the most prevalent malignancies worldwide. No biomarkers that can be used for effective monitoring of therapeutic interventions for this cancer have been identified to date. This study investigated polar metabolite profiles in urine samples from 100 BC patients and 100 normal controls (NCs) using nuclear magnetic resonance (NMR) and two methods of high- resolution nanoparticle-based laser desorption/ionization mass spectrometry (LDI-MS). Five urine metabolites were identified and quantified using NMR spectroscopy to be potential indicators of bladder cancer. Twenty-five LDI-MS-detected compounds, predominantly peptides and lipids, distinguished urine samples from BC and NCs individuals. Level changes of three characteristic urine metabolites enabled BC tumor grades to be distinguished, and ten metabolites were reported to correlate with tumor stages. Receiver-Operating Characteristics analysis showed high predictive power for all three types of metabolomics data, with the area under the curve (AUC) values greater than 0.87. These findings suggest that metabolite markers identified in this study may be useful for the non-invasive detection and monitoring of bladder cancer stages and grades.
more »
« less
Noninvasive diagnostic imaging using machine-learning analysis of nanoresolution images of cell surfaces: Detection of bladder cancer
We report an approach in diagnostic imaging based on nanoscale-resolution scanning of surfaces of cells collected from body fluids using a recent modality of atomic force microscopy (AFM), subresonance tapping, and machine-leaning analysis. The surface parameters, which are typically used in engineering to describe surfaces, are used to classify cells. The method is applied to the detection of bladder cancer, which is one of the most common human malignancies and the most expensive cancer to treat. The frequent visual examinations of bladder (cytoscopy) required for follow-up are not only uncomfortable for the patient but a serious cost for the health care system. Our method addresses an unmet need in noninvasive and accurate detection of bladder cancer, which may eliminate unnecessary and expensive cystoscopies. The method, which evaluates cells collected from urine, shows 94% diagnostic accuracy when examining five cells per patient’s urine sample. It is a statistically significant improvement (
- NSF-PAR ID:
- 10080774
- Publisher / Repository:
- Proceedings of the National Academy of Sciences
- Date Published:
- Journal Name:
- Proceedings of the National Academy of Sciences
- Volume:
- 115
- Issue:
- 51
- ISSN:
- 0027-8424
- Page Range / eLocation ID:
- p. 12920-12925
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Cancer diagnostics is an important field of cancer recovery and survival with many expensive procedures needed to administer the correct treatment. Machine Learning (ML) approaches can help with the diagnostic prediction from circulating tumor cells in liquid biopsy or from a primary tumor in solid biopsy. After predicting the metastatic potential from a deep learning model, doctors in a clinical setting can administer a safe and correct treatment for a specific patient. This paper investigates the use of deep convolutional neural networks for predicting a specific cancer cell line as a tool for label free identification. Specifically, deep learning strategies for weight initialization and performance metrics are described, with transfer learning and the accuracy metric utilized in this work. The equipment used for prediction involves brightfield microscopy without the use of chemical labels, advanced instruments, or time-consuming biological techniques, giving an advantage over current diagnostic methods. In the procedure, three different binary datasets of well-known cancer cell lines were collected, each having a difference in metastatic potential. Two different classification models were adopted (EfficientNetV2 and ResNet-50) with the analysis given for each stage in the ML architecture. The training results for each model and dataset are provided and systematically compared. We found that the test set accuracy showed favorable performance for both ML models with EfficientNetV2 accuracy reaching up to 99%. These test results allowed EfficientNetV2 to outperform ResNet-50 at an average percent increase of 3.5% for each dataset. The high accuracy obtained from the predictions demonstrates that the system can be retrained on a large-scale clinical dataset.
-
more » « less
Abstract Breast cancer is the most common cancer detected in women and current screening methods for the disease are not sensitive. Volatile organic compounds (VOCs) include endogenous metabolites that provide information about health and disease which might be useful to develop a better screening method for breast cancer. The goal of this study was to classify mice with and without tumors and compare tumors localized to the mammary pad and tumor cells injected into the iliac artery by differences in VOCs in urine. After 4T1.2 tumor cells were injected into BALB/c mice either in the mammary pad or into the iliac artery, urine was collected, VOCs from urine headspace were concentrated by solid phase microextraction and results were analyzed by gas chromatography-mass spectrometry quadrupole time-of-flight. Multivariate and univariate statistical analyses were employed to find potential biomarkers for breast cancer and metastatic breast cancer in mice models. A set of six VOCs classified mice with and without tumors with an area under the receiver operator characteristic (ROC AUC) of 0.98 (95% confidence interval [0.85, 1.00]) via five-fold cross validation. Classification of mice with tumors in the mammary pad and iliac artery was executed utilizing a different set of six VOCs, with a ROC AUC of 0.96 (95% confidence interval [0.75, 1.00]).
-
more » « less
In vivo photoacoustic (PA) flow cytometry (PAFC) has great clinical potential for early, noninvasive diagnosis of cancer, infections (e.g., malaria and bacteremia), sickle anemia, and cardiovascular disorders, including stroke prevention through detection of circulating white clots with negative PA contrast. For clinical applications, this diagnostic platform still requires optimization and calibration. We have already demonstrated that this need can be partially addressed byin vivo examination of large mouse blood vessels, which are similar to human vessels used. Here, we present an alternative method for PAFC optimization that utilizes novel, clinically relevant phantoms resembling pigmented skin, tissue, vessels, and flowing blood. This phantom consists of a scattering-absorbing medium with a melanin layer and plastic tube with flowing beads to model light-absorbing red blood cells (RBCs) and circulating tumor cells (CTCs), as well as transparent beads to model white blood cells and clots. Using a laser diode, we demonstrated the extraordinary ability of PAFC to dynamically detect fast-moving mimic CTCs with positive PA contrast and white clots with negative PA contrast in an RBC background. Time-resolved detection of the delayed PA signals from blood vessels demonstrated complete suppression of the PA background from the modeled pigmented skin. This novel, medically relevant, dynamic blood flow phantom can be used to calibrate and maintain PAFC parameters for routine clinical applications. -
more » « less
Advances in medical imaging technology have created opportunities for computer‐aided diagnostic tools to assist human practitioners in identifying relevant patterns in massive, multiscale digital pathology slides. This work presents Hierarchical Linear Time Subset Scanning, a novel statistical method for pattern detection. Hierarchical Linear Time Subset Scanning exploits the hierarchical structure inherent in data produced through virtual microscopy in order to accurately and quickly identify regions of interest for pathologists to review. We take a digital image at various resolution levels, identify the most anomalous regions at a coarse level, and continue to analyze the data at increasingly granular resolutions until we accurately identify its most anomalous subregions. We demonstrate the performance of our novel method in identifying cancerous locations on digital slides of prostate biopsy samples and show that our methods detect regions of cancer in minutes with high accuracy, both as measured by the ROC curve (measuring ability to distinguish between benign and cancerous slides) and by the spatial precision‐recall curve (measuring ability to pick out the malignant areas on a slide which contains cancer). Existing methods need small scale images (small areas of a slide preselected by the pathologist for analysis, eg, 32 × 32 pixels) and may not work effectively on large, raw digitized images of size 100
K × 100K pixels. In this work, we provide a methodology to fill this significant gap by analyzing large digitized images and identifying regions of interest that may be indicative of cancer.
