An official website of the United States government
Infrared breast thermography has been associated with the early detection of breast cancer. However, findings in previous studies have been inconclusive. The upright position of subjects during imaging introduces errors in interpretation because it blocks the optical access in the inframammary fold region and alters the temperature due to contact between breast and chest wall. These errors can be avoided by imaging breasts in prone position. Although the numerical simulations provide insight into thermal characteristics of the female breast with a tumor, most simulations in the past have used cubical and hemispherical breast models. We hypothesize that a breast model with the actual breast shape will provide true thermal characteristics that are useful in tumor detection. A digital breast model in prone position is developed to generate the surface temperature profiles for breasts with tumors. The digital breast model is generated from sequential MRI images and simulations are performed using Finite Volume Method employing Pennes bioheat equation. We investigated the effect of varying the tumor metabolic activity on the surface temperature profile. We compared the surface temperature profile for various tumor metabolic activities with a case without tumor. The resulting surface temperature rise near the location of the tumor was between 0.665 and 1.023 °C, detectable using modern Infrared cameras. This is the first time that numerical simulations are conducted in a model with the actual breast shape in prone position to study the surface temperature changes induced by breast cancer.
more »
« less
Thermal Modeling of Patient-Specific Breast Cancer With Physics-Based Artificial Intelligence
Abstract Breast cancer is a prevalent form of cancer among women. It is associated with increased heat generation due to higher metabolism in the tumor and increased blood vessels resulting from angiogenesis. The thermal alterations result in a change in the breast surface temperature profile. Infrared imaging is an FDA-approved adjunctive to mammography, which employs the surface temperature alterations in detecting cancer. To apply infrared imaging in clinical settings, it is necessary to develop effective techniques to model the relation between the tumor characteristics and the breast surface temperatures. The present work describes the thermal modeling of breast cancer with physics-informed neural networks. Losses are assigned to random points in the domain based on the boundary conditions and governing equations that should be satisfied. The Adam optimizer in TensorFlow minimizes the losses to find the temperature field or thermal conductivity that satisfies the boundary conditions and the bioheat equation. Backpropagation computes the derivatives in the bioheat equation. Analyses of the three patient-specific cases show that the machine-learning model accurately reproduces the thermal behavior given by ansys-fluent simulation. Also, good agreement between the model prediction and the infrared images is observed. Moreover, the neural network accurately recovers the thermal conductivity within 6.5% relative error.
more »
« less
- Award ID(s):
- 2136325
- PAR ID:
- 10391187
- Date Published:
- Journal Name:
- Journal of Heat Transfer
- Volume:
- 145
- Issue:
- 3
- ISSN:
- 2832-8450
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Early and accurate detection of breast cancer is a critical part of the strategy to reduce the morbidity and mortality associated with this common disease. While current guidelines recommend mammography for screening, the sensitivity and specificity of mammograms remains less than optimal, especially for patients with dense breast tissue. Thermography has been explored in the past as an alternative to mammography. Advances in IR cameras that are used to obtain thermal images of the breast as well as computational tools used to accurately model heat transfer within the breast have significantly increased the accuracy of thermography. The current work reviews the progress that has been made in using thermal imaging to detect breast cancer over the past three decades and identifies aspects that need further refinement for it to become a reliable tool to diagnose breast cancer. Recent advances and suggestions for future work in the field including using advanced simulation methods, inverse modeling, imaging protocols, and using artificial neural networks to better predict the location of the tumor are also presented.more » « less
-
Abstract Breast cancer metastasis occurs via blood and lymphatic vessels. Breast cancer cells ‘educate’ lymphatic endothelial cells (LECs) to support tumor vascularization and growth. However, despite known metabolic alterations in breast cancer, it remains unclear how lymphatic endothelial cell metabolism is altered in the tumor microenvironment and its effect in lymphangiogenic signaling in LECs. We analyzed metabolites inside LECs in co-culture with MCF-7, MDA-MB-231, and SK-BR-3 breast cancer cell lines using $$^1\hbox {H}$$ 1 H nuclear magnetic resonance (NMR) metabolomics, Seahorse, and the spatial distribution of metabolic co-enzymes using optical redox ratio imaging to describe breast cancer-LEC metabolic crosstalk. LECs co-cultured with breast cancer cells exhibited cell-line dependent altered metabolic profiles, including significant changes in lactate concentration in breast cancer co-culture. Cell metabolic phenotype analysis using Seahorse showed LECs in co-culture exhibited reduced mitochondrial respiration, increased reliance on glycolysis and reduced metabolic flexibility. Optical redox ratio measurements revealed reduced NAD(P)H levels in LECs potentially due to increased NAD(P)H utilization to maintain redox homeostasis. $$^{13}\hbox {C}$$ 13 C -labeled glucose experiments did not reveal lactate shuttling into LECs from breast cancer cells, yet showed other $$^{13}\hbox {C}$$ 13 C signals in LECs suggesting internalized metabolites and metabolic exchange between the two cell types. We also determined that breast cancer co-culture stimulated lymphangiogenic signaling in LECs, yet activation was not stimulated by lactate alone. Increased lymphangiogenic signaling suggests paracrine signaling between LECs and breast cancer cells which could have a pro-metastatic role.more » « less
-
Mesenchymal stem cells (MSCs) that accumulate in the primary tumor due to their natural tropism for inflammatory tissues enhance the metastatic potential of tumor cells through direct interactions with tumor cells or paracrine signaling within the tumor microenvironment. MSCs also undergo senescence, which leads to increased production of pro-inflammatory cytokines and matrix-degrading enzymes. Senescence is a critical mechanism of limiting abnormal growth and cancer development through tumor suppression; however, senescent cells that accumulate in tissues eventually develop a senescence-associated secretory phenotype that alters the microenvironment to promote cancer. Increased understanding of the biophysical properties of senescent MSCs and how they mediate cell-cell interactions in the tumor may be useful in identifying novel biomarkers for senescent stromal cells in tissues or aggressive cancer cells that form in an aging stroma. A high-content single cell biophysical approach was used to define the mechanical properties of pre- and post- senescent MSCs. Our data shows post-senescent MSCs are larger and less motile, with more homogeneous mechanical properties than their pre-senescent counterparts. A robust molecular screening approach combining genome-wide microarray analysis with mass spec-based proteomics was used to establish the molecular differences in pre- and post- senescent MSCs. Our data show a consistent correlation of up and down regulated gene and peptide expression. A 3D co-culture model was used to assess the effects of pre- and post- senescent MSCs on breast cancer cell motility and invasion in 3D collagen gels. Post-senescent MSCs induced an invasive breast cancer cell phenotype, characterized by increased spreading of breast cancer cells in collagen, increased numbers of invading cells, and morphological elongation of breast cancer cells. Surprisingly, this invasive breast cancer cell behavior was further amplified when breast cancer cells were co-cultured with both pre- and post- senescent cells.more » « less
-
A device for measuring a plurality of material properties is designed to include accurate sensors configured to consecutively obtain thermal conductivity, electrical conductivity, and Seebeck coefficient of a single sample while maintaining a vacuum or inert gas environment. Four major design factors are identified as sample-heat spreader mismatch, radiation losses, parasitic losses, and sample surface temperature variance. The design is analyzed using finite element methods for high temperature ranges up to 1000°C as well as ultra-high temperatures up to 2500°C. A temperature uncertainty of 0.46% was estimated for a sample with cold and hot sides at 905.1 and 908.5°C, respectively. The uncertainty at 1000°C was calculated to be 0.7% for a ?T of 5°C between the hot and cold sides. The thermal conductivity uncertainty was calculated to be -8.6% at ~900°C for a case with radiative gains, and +8.2% at ~1000°C for a case with radiative losses, indicating the sensitivity of the measurement to the temperature of the thermal guard in relation to the heat spreader and sample temperature. Lower limits of -17 and -13% error in thermal conductivity measurements were estimated at the ultra-high temperature of ~2500°C for a single-stage and double-stage radiation shield, respectively. It is noted that this design is not limited to electro-thermal characterization and will enable measurement of ionic conductivity and surface temperatures of energy materials under realistic operating conditions in extreme temperature environments.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1115/1.4055347
