1
|
Dank G, Buber T, Rice A, Kraicer N, Hanael E, Shasha T, Aviram G, Yehudayoff A, Kent MS. Training and validation of a novel non-invasive imaging system for ruling out malignancy in canine subcutaneous and cutaneous masses using machine learning in 664 masses. Front Vet Sci 2023; 10:1164438. [PMID: 37841459 PMCID: PMC10570610 DOI: 10.3389/fvets.2023.1164438] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/12/2023] [Accepted: 09/05/2023] [Indexed: 10/17/2023] Open
Abstract
Objective To train and validate the use of a novel artificial intelligence-based thermal imaging system as a screening tool to rule out malignancy in cutaneous and subcutaneous masses in dogs. Animals Training study: 147 client-owned dogs with 233 masses. Validation Study: 299 client-owned dogs with 525 masses. Cytology was non-diagnostic in 94 masses, resulting in 431 masses from 248 dogs with diagnostic samples. Procedures The prospective studies were conducted between June 2020 and July 2022. During the scan, each mass and its adjacent healthy tissue was heated by a high-power Light-Emitting Diode. The tissue temperature was recorded by the device and consequently analyzed using a supervised machine learning algorithm to determine whether the mass required further investigation. The first study was performed to collect data to train the algorithm. The second study validated the algorithm, as the real-time device predictions were compared to the cytology and/or biopsy results. Results The results for the validation study were that the device correctly classified 45 out of 53 malignant masses and 253 out of 378 benign masses (sensitivity = 85% and specificity = 67%). The negative predictive value of the system (i.e., percent of benign masses identified as benign) was 97%. Clinical relevance The results demonstrate that this novel system could be used as a decision-support tool at the point of care, enabling clinicians to differentiate between benign lesions and those requiring further diagnostics.
Collapse
Affiliation(s)
| | - Tali Buber
- HT BioImaging Ltd., Hod Hasharon, Israel
| | - Anna Rice
- HT BioImaging Ltd., Hod Hasharon, Israel
| | | | | | | | - Gal Aviram
- Department Biomedical Engineering, Tel Aviv University, Tel Aviv-Yafo, Israel
| | | | - Michael S. Kent
- Department of Surgical and Radiological Sciences, School of Veterinary Medicine, University of California, Davis, Davis, CA, United States
| |
Collapse
|
2
|
Dank G, Buber T, Polliack G, Aviram G, Rice A, Yehudayoff A, Kent MS. A pilot study for a non-invasive system for detection of malignancy in canine subcutaneous and cutaneous masses using machine learning. Front Vet Sci 2023; 10:1109188. [PMID: 36777665 PMCID: PMC9909829 DOI: 10.3389/fvets.2023.1109188] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/27/2022] [Accepted: 01/09/2023] [Indexed: 01/27/2023] Open
Abstract
Introduction Early diagnosis of cancer enhances treatment planning and improves prognosis. Many masses presenting to veterinary clinics are difficult to diagnose without using invasive, time-consuming, and costly tests. Our objective was to perform a preliminary proof-of-concept for the HT Vista device, a novel artificial intelligence-based thermal imaging system, developed and designed to differentiate benign from malignant, cutaneous and subcutaneous masses in dogs. Methods Forty-five dogs with a total of 69 masses were recruited. Each mass was clipped and heated by the HT Vista device. The heat emitted by the mass and its adjacent healthy tissue was automatically recorded using a built-in thermal camera. The thermal data from both areas were subsequently analyzed using an Artificial Intelligence algorithm. Cytology and/or biopsy results were later compared to the results obtained from the HT Vista system and used to train the algorithm. Validation was done using a "Leave One Out" cross-validation to determine the algorithm's performance. Results The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of the system were 90%, 93%, 88%, 83%, and 95%, respectively for all masses. Conclusion We propose that this novel system, with further development, could be used to provide a decision-support tool enabling clinicians to differentiate between benign lesions and those requiring additional diagnostics. Our study also provides a proof-of-concept for ongoing prospective trials for cancer diagnosis using advanced thermodynamics and machine learning procedures in companion dogs.
Collapse
Affiliation(s)
- Gillian Dank
- Koret School of Veterinary Medicine, Hebrew University, Rehovot, Israel,*Correspondence: Gillian Dank ✉
| | | | | | - Gal Aviram
- HT BioImaging LTD, Hod Hasharon, Israel,Department Biomedical Engineering, Tel Aviv University, Tel Aviv, Israel
| | - Anna Rice
- HT BioImaging LTD, Hod Hasharon, Israel
| | | | - Michael S. Kent
- Department of Surgical and Radiological Sciences, School of Veterinary Medicine, University of California, Davis, Davis, CA, United States
| |
Collapse
|
3
|
Ehrlich LE, Fahy GM, Wowk BG, Malen JA, Rabin Y. Thermal Analyses of a Human Kidney and a Rabbit Kidney During Cryopreservation by Vitrification. J Biomech Eng 2018; 140:2646923. [PMID: 28753690 PMCID: PMC5676646 DOI: 10.1115/1.4037406] [Citation(s) in RCA: 18] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/14/2017] [Revised: 07/07/2017] [Indexed: 12/12/2022]
Abstract
This study focuses on thermal analysis of the problem of scaling up from the vitrification of rabbit kidneys to the vitrification of human kidneys, where vitrification is the preservation of biological material in the glassy state. The basis for this study is a successful cryopreservation protocol for a rabbit kidney model, based on using a proprietary vitrification solution known as M22. Using the finite element analysis (FEA) commercial code ANSYS, heat transfer simulations suggest that indeed the rabbit kidney unquestionably cools rapidly enough to be vitrified based on known intrarenal concentrations of M22. Scaling up 21-fold, computer simulations suggest less favorable conditions for human kidney vitrification. In this case, cooling rates below -100 °C are sometimes slower than 1 °C/min, a rate that provides a clear-cut margin of safety at all temperatures based on the stability of rabbit kidneys in past studies. Nevertheless, it is concluded in this study that vitrifying human kidneys is possible without significant ice damage, assuming that human kidneys can be perfused with M22 as effectively as rabbit kidneys. The thermal analysis suggests that cooling rates can be further increased by a careful design of the cryogenic protocol and by tailoring the container to the shape of the kidney, in contrast to the present cylindrical container. This study demonstrates the critical need for the thermal analysis of experimental cryopreservation and highlights the unmet need for measuring the thermophysical properties of cryoprotective solutions under conditions relevant to realistic thermal histories.
Collapse
Affiliation(s)
- Lili E. Ehrlich
- Department of Mechanical Engineering,
Carnegie Mellon University,
Pittsburgh, PA 15213
| | | | | | - Jonathan A. Malen
- Department of Mechanical Engineering,
Carnegie Mellon University,
Pittsburgh, PA 15213
| | - Yoed Rabin
- Department of Mechanical Engineering,
Carnegie Mellon University,
Pittsburgh, PA 15213
e-mail:
| |
Collapse
|
4
|
Chan JY, Ooi EH. Sensitivity of thermophysiological models of cryoablation to the thermal and biophysical properties of tissues. Cryobiology 2016; 73:304-315. [DOI: 10.1016/j.cryobiol.2016.10.006] [Citation(s) in RCA: 8] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/03/2016] [Revised: 10/18/2016] [Accepted: 10/20/2016] [Indexed: 10/20/2022]
|
5
|
Bruggmoser G, Bauchowitz S, Canters R, Crezee H, Ehmann M, Gellermann J, Lamprecht U, Lomax N, Messmer M, Ott O, Abdel-Rahman S, Schmidt M, Sauer R, Thomsen A, Wessalowski R, van Rhoon G. Guideline for the clinical application, documentation and analysis of clinical studies for regional deep hyperthermia. Strahlenther Onkol 2012; 188 Suppl 2:198-211. [DOI: 10.1007/s00066-012-0176-2] [Citation(s) in RCA: 74] [Impact Index Per Article: 6.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/04/2023]
|
6
|
Delhomme G, Newman WH, Roussel B, Jouvet M, Bowman HF, Dittmar A. Thermal diffusion probe and instrument system for tissue blood flow measurements: validation in phantoms and in vivo organs. IEEE Trans Biomed Eng 1994; 41:656-62. [PMID: 7927386 DOI: 10.1109/10.301732] [Citation(s) in RCA: 16] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/27/2023]
Abstract
A minimally invasive probe and instrument system for real-time measurements of temperature, thermal conductivity and tissue blood flow has been designed for research and clinical use. The essence of the probe is a thermistor, located at the tip of catheters or glass and steel needles, and operating in transient self-heated mode at constant temperature increment. Thermal conductivity and tissue blood flow are determined by use of a coupled tissue-probe thermal model. The effects of temporal baseline temperature shifts are minimized by a novel, automatic, analog compensation circuit. Very short heating periods (3 s) and cooling periods (12 s) provided near-continuous measurements (4/min). Calibration experiments performed in media of known thermal conductivity exhibit a linear response with respect to thermal conductivity. In vitro experiments performed in isolated perfused dog liver preparations are presented to evaluate this instrument system. In vivo experiments performed in cat brain, dog liver, and human tumor demonstrate the ability of this instrument system to perform physiologically valid measurements (comparison inter-subjects and intra-subjects). The minimally invasive probes (0.8 mm OD) are capable of long term measurements (several months), with minimal tissue reactions (0.3 mm around the probe).
Collapse
Affiliation(s)
- G Delhomme
- Laboratoire de Thermorégulation, URA 1341 CNRS, Faculté de médecine, Lyon, France
| | | | | | | | | | | |
Collapse
|
7
|
Brown SL, Li XL, Pai HH, Worthington AE, Hill RP, Hunt JW. Observations of thermal gradients in perfused tissues during water bath heating. Int J Hyperthermia 1992; 8:275-87. [PMID: 1573316 DOI: 10.3109/02656739209021782] [Citation(s) in RCA: 11] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/27/2022] Open
Abstract
Actual thermal gradients in perfused tissues are difficult to observe using thermocouples because of thermal conduction along the probes. We have used fine type-K (chromel-alumel) probes, which have a much lower thermal conductivity than equivalent-sized type-T (copper-constantan) thermocouples, to examine thermal gradients in two mouse tumour systems during water bath heating. The results indicate substantial heterogeneity in temperature distribution even in tumours transplanted in the foot and immersed to a depth of 2 cm in a 44 degrees C water bath for 20 min, i.e. thermal gradients greater than 1 degree C/mm were observed in KHT fibrosarcomas. The temperature heterogeneity for water bath heating is primarily a result of blood flow and appears to be tumour-specific. Temperature measurements using an excised perfused canine kidney demonstrate that increased perfusate volume flow increases the range of tissue temperatures. Consistent with theory, an artifactual improvement in temperature homogeneity resulted when temperature was measured using type-T thermocouples instead of type-K probes. These results emphasize the difficulties in obtaining accurate temperature measurements during experimental and clinical hyperthermia. Even extensive measurements of temperature in tissues may underestimate the true range of heterogeneity unless factors such as thermal smearing are controlled.
Collapse
MESH Headings
- Animals
- Body Temperature
- Carcinoma, Squamous Cell/blood supply
- Carcinoma, Squamous Cell/physiopathology
- Carcinoma, Squamous Cell/therapy
- Dogs
- Evaluation Studies as Topic
- Hyperthermia, Induced/methods
- Immersion
- Mice
- Mice, Inbred C3H
- Neoplasms, Experimental/blood supply
- Neoplasms, Experimental/physiopathology
- Neoplasms, Experimental/therapy
- Sarcoma, Experimental/blood supply
- Sarcoma, Experimental/physiopathology
- Sarcoma, Experimental/therapy
- Thermometers
- Water
Collapse
Affiliation(s)
- S L Brown
- Ontario Cancer Institute, Toronto, Canada
| | | | | | | | | | | |
Collapse
|
8
|
Benkeser PJ, Frizzell LA, Holmes KR, Goss SA. A perfused tissue phantom, for ultrasound hyperthermia. IEEE Trans Biomed Eng 1990; 37:425-8. [PMID: 2338357 DOI: 10.1109/10.52351] [Citation(s) in RCA: 12] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/31/2022]
Abstract
A perfused tissue phantom, developed as a tool for analyzing the performance of ultrasound hyperthermia applicators, was investigated. The phantom, consisting of a fixed porcine kidney with thermocouples placed throughout the tissue, was perfused with degassed water by a variable flow rate pump. The phantom was insonated by an unfocused multielement ultrasound applicator and the temperatures in the phantom were recorded. The results indicate that for testing protocols where tissue phantoms are needed, the fixed kidney preparation offers an opportunity to use a more realistic phantom than has previously been available to assess the heating performance of ultrasound hyperthermia applicators.
Collapse
Affiliation(s)
- P J Benkeser
- School of Electrical Engineering, Georgia, Institute of Technology, Atlanta 30332
| | | | | | | |
Collapse
|
9
|
Adams T, Spielman WS, Holmes KR, Heisey SR, Chen MM. Proposed methods for the measurement of regional renal blood flow using heat transfer analysis. Ann Biomed Eng 1985; 13:237-58. [PMID: 3898927 DOI: 10.1007/bf02584242] [Citation(s) in RCA: 8] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/07/2023]
Abstract
The kidney, with its heterogeneous regional perfusion in the two anatomically and functionally distinct vascular beds of the renal cortex and medulla, and with its non-uniform blood vessel geometries, presents a unique challenge for measuring intrarenal blood flow distribution. Determining whole organ perfusion, on the other hand, is comparatively simple for the kidney, but it provides relatively little information about the suspected dependency of renal excretory function on local perfusion rate. Among the variety of methods proposed for gauging regional renal blood flow, some depend on measuring one or more of the tissue's thermal properties. The most straightforward, but least reliable, involve measurements either of focal tissue temperature alone, or of regional tissue thermal gradients. Simply using heat as a diffusible indicator, however, is unreliable as a measure of blood flow, for many of the same reasons that using an inert gas in a dilution technique is unreliable. Recently developed thermal analytical methods, though, hold promise for measuring local tissue blood flow with accuracy and precision. Two of them are reviewed here. One depends on measurement of the effective thermal conductivity of a small mass of tissue by evaluating the steady state ratio between regional unidirectional heat flux across it and the associated temperature gradient in one vector along a segment of it through an imposed spheroidal heat field. The other depends on analyses of tissue temperature decay subsequent to a controlled pulse of heat delivered through a small inserted thermistor bead. Both techniques use bioheat transfer equations to deduce regional blood flow by differentiating between heat dissipation due to local thermal conductivity and that attributable to the effects of regional convection. Although both methods are unavoidably invasive, neither produces debilitating damage in the tissue volume in which perfusion is measured, nor increases local temperature or metabolism enough to affect blood flow itself. Both techniques quantify local blood flow in small volumes of tissue by detailed evaluation of the many properties of tissue and blood which affect heat transfer, and both allow for a virtually unlimited number of nearly continuous sequential measurements at short (nom. 1 min) time intervals.
Collapse
|