RESUMO

Infrared thermography has been investigated in recent studies to monitor body surface temperature and correlate it with animal welfare and performance factors. In this context, this study proposes the use of the thermal signature method as a feature extractor from the temperature matrix obtained from regions of the body surface of laying hens (face, eye, wattle, comb, leg, and foot) to enable the construction of a computational model for heat stress level classification. In an experiment conducted in climate-controlled chambers, 192 laying hens, 34 weeks old, from two different strains (Dekalb White and Dekalb Brown) were divided into groups and housed under conditions of heat stress (35 °C and 60% humidity) and thermal comfort (26 °C and 60% humidity). Weekly, individual thermal images of the hens were collected using a thermographic camera, along with their respective rectal temperatures. Surface temperatures of the six featherless image areas of the hens' bodies were cut out. Rectal temperature was used to label each infrared thermography data as "Danger" or "Normal", and five different classifier models (Random Forest, Random Tree, Multilayer Perceptron, K-Nearest Neighbors, and Logistic Regression) for rectal temperature class were generated using the respective thermal signatures. No differences between the strains were observed in the thermal signature of surface temperature and rectal temperature. It was evidenced that the rectal temperature and the thermal signature express heat stress and comfort conditions. The Random Forest model for the face area of the laying hen achieved the highest performance (89.0%). For the wattle area, a Random Forest model also demonstrated high performance (88.3%), indicating the significance of this area in strains where it is more developed. These findings validate the method of extracting characteristics from infrared thermography. When combined with machine learning, this method has proven promising for generating classifier models of thermal stress levels in laying hen production environments.

RESUMO

OBJECTIVE: Conical orthopedic drill bits may have the potential to improve the stabilization of orthopedic screws. During perforations, heat energy is released, and elevated temperatures could be related to thermal osteonecrosis. This study was designed to evaluate the thermal behavior of an experimental conical drill bit, when compared to the conventional cylindrical drill, using polyurethane blocks perforations. RESULTS: The sample was divided into two groups, according to the method of drilling, including 25 polyurethane blocks in each: In Group 1, perforations were performed with a conventional orthopedic cylindrical drill; while in Group 2, an experimental conical drill was used. No statistically significant difference was observed in relation to the maximum temperature (MT) during the entire drilling in the groups, however the perforation time (PT) was slightly longer in Group 2. Each drill bit perforated five times and number of perforations was not correlated with a temperature increase, when evaluated universally or isolated by groups. The PT had no correlation with an increase in temperature when evaluating the perforations universally (n = 50) and in Group 1 alone; however, Group 2 showed an inversely proportional correlation for these variables, indicating that, for the conical drill bit, drillings with longer PT had lower MT.

Assuntos

RESUMO

Skin temperature responses have been advocated to indicate exercise-induced muscle soreness and recovery status. While the evidence is contradictory, we hypothesize that the presence of muscle damage and the time window of measurement are confounding factors in the skin temperature response. The objective was to determine whether skin temperature is influenced by different workloads and the time course of temperature measurements over the following 24 h. 24 trained male military were assigned to one of three groups: GC group (n = 8) serving as control not performing exercises, GE group (n = 8) performing a simulated military combat protocol in an exercise track with different obstacles but designed not to elicit muscle damage, and the GEMD group (n = 8) performing the simulated military combat protocol plus 5 sets of 20 drop jumps, with 10-sec between repetitions and with 2-min of rest between sets aiming to induce muscle damage. Skin temperature was measured using infrared thermography before exercise (Pre) and 4 (Post4h), 8 (Post8h) and 24h (Post24h) post-exercise. Perception of pain (DOMS) was evaluated Pre, Post24h, and Post48h, and countermovement jump height was evaluated at Pre and Post24h. DOMS did not differ between groups in the Pre and Post24h measures but GEMD presented higher DOMS than the other groups at Post48h (p < 0.001 and large effect size). Jump height did not differ for GEMD and GC, and GE presented higher jump height at Post24h than GC (p = 0.02 and large effect size). Skin temperature responses of GEMD and GG were similar in all measurement moments (p > 0.22), and GE presented higher skin temperature than the GC and the GEMD groups at Post24h (p < 0.01 and large effect sizes). In conclusion, although physical exercise elicits higher skin temperature that lasts up to 24 h following the efforts, muscle soreness depresses this response.

Assuntos

RESUMO

Infrared thermography is gaining relevance in breast cancer assessment. For this purpose, breast segmentation in thermograms is an important task for performing automatic image analysis and detecting possible temperature changes that indicate the presence of malignancy. However, it is not a simple task since the breast limit borders, especially the top borders, often have low contrast, making it difficult to isolate the breast area. Several algorithms have been proposed for breast segmentation, but these highly depend on the contrast at the lower breast borders and on filtering algorithms to remove false edges. This work focuses on taking advantage of the distinctive inframammary shape to simplify the definition of the lower breast border, regardless of the contrast level, which indeed also provides a strong anatomical reference to support the definition of the poorly marked upper boundary of the breasts, which has been one of the major challenges in the literature. In order to demonstrate viability of the proposed technique for an automatic breast segmentation, we applied it to a database with 180 thermograms and compared their results with those reported by others in the literature. We found that our approach achieved a high performance, in terms of Intersection over Union of 0.934, even higher than that reported by artificial intelligence algorithms. The performance is invariant to breast sizes and thermal contrast of the images.

Assuntos

RESUMO

Most of the responses present in animals when exposed to stressors are mediated by the autonomic nervous system. The sympathetic nervous system, known as the one responsible for the "fight or flight" reaction, triggers cardiovascular changes such as tachycardia or vasomotor alterations to restore homeostasis. Increase in body temperature in stressed animals also activates peripheral compensatory mechanisms such as cutaneous vasodilation to increase heat exchange. Since changes in skin blood flow influence the amount of heat dissipation, infrared thermography is suggested as a tool that can detect said changes. The present review aims to analyze the application of infrared thermography as a method to assess stress-related autonomic activity, and their association with the cardiovascular and heart rate variability in domestic animals.

RESUMO

OBJECTIVE: Accurate assessment of burn depth and burn wound healing potential is essential to determine early treatments. Infrared thermography (IRT) is a non-invasive and objective tool to do this. This systematic review evaluated the accuracy of IRT to determine burn wound healing potential. METHOD: This systematic review and meta-analysis used MEDLINE, EMBASE, CINAHL, PEDro, DiTA and CENTRAL databases. IRT data were extracted from primary studies and categorised into four cells (i.e., true positives, false positives, true negatives and false negatives). Subgroup analysis was performed according to methods used to capture thermal images. RESULTS: The search strategy identified 2727 publications; however, 15 articles were selected for review and 11 for meta-analysis. In our meta-analysis, the accuracy of IRT was 84.8% (63% sensitivity and 81.9% specificity). CONCLUSION: IRT is a moderately accurate tool to identify burn depth and healing potential. Thus, IRT should be used carefully for evaluating burn wounds.

Assuntos

RESUMO

Infrared Thermography (IRT) has become an assistance tool in medicine and is used to noninvasively evaluate heat elimination during and after inflammatory processes or during the recovery period. However, its application in veterinary patients undergoing physiotherapy is a field that requires deep research. This review aims to analyze the application of IRT in the monitoring of animal physiotherapy, using the thermal changes that are present in patients undergoing gait or lameness issues (e.g., inflammation, pain, increased local temperature) as a neurobiological basis. Rehabilitation techniques such as acupuncture, physical therapies, thermotherapy, photo-biomodulation, and electrostimulation have been reported to have an anti-inflammatory effect that decreases the amount of local heat production, which is heat that can be recorded with IRT. Therefore, IRT could be used as a complementary tool to evaluate the effectiveness of the therapy, and it is suggested that further studies evaluate the accuracy, sensibility, and sensitivity of IRT.

RESUMO

Background: Epilepsy surgery represents a therapeutic opportunity for those patients who do not respond to drug therapy. However, an important challenge is the precise identification of the epileptogenic area during surgery. Since it can be hard to delineate, it makes it necessary to use auxiliary tools as a guide during the surgical procedure. Electrocorticography (ECoG), despite having shown favorable results in terms of reducing post-surgical seizures, have certain limitations. Brain mapping using infrared thermography mapping and a new thermosensitive/thermochromic silicone (TTS) in epilepsy surgery has introduced a new resource of noninvasive and real-time devices that allow the localization of irritative zones. Methods: Sixty consecutive patients with drug-resistant epilepsy with surgical indications who decided to participate voluntarily in the study were included in the study. We measured brain temperature using two quantitative methods and a qualitative method: the TTS sheet. In all cases, we used ECoG as the gold standard to identify irritative areas, and all brain tissue samples obtained were sent to pathology for diagnosis. Results: In the subgroup in which the ECoG detected irritative areas (n = 51), adding the results in which there was a correlation with the different methods, the efficiency obtained to detect irritative areas is 94.11% (n = 48/51, P ≤ 0.0001) while the infrared thermography mapping method independently has an efficiency of 91.66% (P ≤ 0.0001). The TTS has a sensitivity of 95.71% and a specificity of 97.9% (P ≤ 0.0001) to detect hypothermic areas that correlate with the irritative zones detected by ECoG. No postoperative infections or wound dehiscence were documented, so the different methodologies used do not represent an additional risk for the surgical proceedings. Conclusion: We consider that the infrared thermography mapping using high-resolution infrared thermography cameras and the TTS are both accurate and safe methods to identify irritative areas in epilepsy surgeries.

RESUMO

When an organism detects decreases in their core body temperature, the hypothalamus, the main thermoregulatory center, triggers compensatory responses. These responses include vasomotor changes to prevent heat loss and physiological mechanisms (e.g., shivering and non-shivering thermogenesis) for heat production. Both types of changes require the participation of peripheral thermoreceptors, afferent signaling to the spinal cord and hypothalamus, and efferent pathways to motor and/or sympathetic neurons. The present review aims to analyze the scientific evidence of the hypothalamic control of hypothermia and the central and peripheral changes that are triggered in domestic animals.

RESUMO

Introduction Infrared thermography (IRT) has demonstrated high diagnostic accuracy for carpal tunnel syndrome (CTS) diagnosis in previous studies. However, the recovery of the autonomic function after treatment for CTS is rarely addressed in the literature, especially on the long-term. Case Presentation A 59-year-old lady sought treatment for a long-term history of numbness, tingling, and hand and arm pain. CTS was diagnosed by clinical and electrophysiological means. After 6 months of conservative treatment, surgical treatment was offered. Preoperative IRT was performed by static and dynamic evaluations immediately and 5 minutes after the cold challenge test using the FLIR C2 camera with accuracy of 2°C or 2%. Fingers were consistently colder (mean of 3.76° C), which clearly represented an autonomic dysfunction in the patient's hand. The patient underwent mini-open carpal tunnel decompression and did great postoperatively. One year after surgery, the patient was fully recovered and completely asymptomatic. IRT imaging showed a remarkable improvement of fingers temperature (mean of 3.36°C). Conclusion Our long-term results confirmed that functional recovery occurred concomitantly to autonomic recovery, which was demonstrated by consistent improvement in fingers' temperature. IRT has a strong potential at the evaluation of patients with CTS for both diagnosis and follow-up.

Introdução A termografia infravermelha (IRT) demonstrou alta precisão diagnóstica para o diagnóstico da síndrome do túnel do carpo (STC) em estudos anteriores. No entanto, a recuperação da função autonômica após o tratamento para STC é raramente abordada na literatura, especialmente a longo prazo. Apresentação do caso Uma senhora de 59 anos procurou tratamento para um histórico de longo prazo de dormência, formigamento e dor nas mãos e braços. A STC foi diagnosticada por meios clínicos e eletrofisiológicos. Após 6 meses de tratamento conservador, o tratamento cirúrgico foi oferecido. A IRT pré-operatória foi realizada por avaliações estáticas e dinâmicas imediatamente e 5 minutos após o teste de provocação pelo frio usando a câmera FLIR C2 com precisão de 2 °C ou 2%. Os dedos estavam consistentemente mais frios (média de 3,76 °C), o que claramente representava uma disfunção autonômica na mão da paciente. A paciente foi submetida a uma mini descompressão aberta do túnel do carpo e teve um ótimo desempenho no pós-operatório. Um ano após a cirurgia, a paciente estava totalmente recuperada e completamente assintomática. A imagem IRT mostrou uma melhora notável na temperatura dos dedos (média de 3,36 °C). Conclusão Nossos resultados de longo prazo confirmaram que a recuperação funcional ocorreu concomitantemente à recuperação autonômica, o que foi demonstrado pela melhora consistente na temperatura dos dedos. A IRT tem um forte potencial na avaliação de pacientes com STC para diagnóstico e acompanhamento.

RESUMO

OBJECTIVES: To evaluate the clinical and thermographic aspects of the gingival phenotype (GP) in healthy subjects. MATERIALS AND METHODS: The study sample examined 264 teeth, upper and lower incisors, comprising 132 central incisors (CI) and 132 lateral incisors (LI), in 33 healthy volunteers. Four periodontal parameters were recorded systematically: ratio of width to length of the dental crown (CW/CL), height of the gingival attachment (KGW), probing depth (PD), and gingival transparency (GT). The temperatures of the attached gingiva (KGW temp) and the free gingival margin (FGM Temp) were also recorded by way of infrared thermography (IRT). RESULTS: The average age of the sample was 30.70 ± 7.65 years. Of the 264 teeth evaluated, 76.1% had a thin GP. There was a significant association between CW/CL (p < 0.001), KGW (p < 0.001), PD (p < 0.007), and FGM Temp (p < 0.006) with the tooth groups. The results show a significant and inversely proportional correlation between clinical parameters and gingival temperature (p < 0.05). A significant association was found between CW/CL (p < 0.026); KGW (p < 0.001); and GP. CONCLUSION: CW/CL, KGW, PD, and FGM Temp vary according to tooth group. The majority of the sample presented with the thin GP, which was more prevalent in the lower LIs and CIs. No significant association could be observed between thermographic parameters and GP. CLINICAL RELEVANCE: Knowledge of the temperature of the gingival tissues can be useful in helping to diagnose and plan periodontal treatments. Moreover, our findings will help future studies evaluate the use of IR as an auxiliary diagnostic method in dentistry, eliminating GP as a confounding factor, since it does not seem to influence the temperature of the gingival tissues.

Assuntos

RESUMO

Transport is a stressor that can cause physiological and metabolic imbalances in livestock, resulting in stress-induced hyperthermia. In water buffaloes, studies regarding the thermal state of animals during mobilization are scarce. Therefore, this study aimed to compare the thermal response of 1516 water buffaloes using infrared thermography (IRT) during 15 short trips (783 animals, 60,291 records, average duration = 50.33 min ± 5.48 min) and 14 long trips (733 animals, 56,441 records, average duration = 13.31 h ± 47.32 min). The surface temperature was assessed in 11 regions (periocular, lacrimal caruncle, nasal, lower eyelid, auricular, frontal-parietal, pelvic limb, torso, abdominal, lumbar, and thoracic) during seven phases from pasture to post-transport. It was found that the surface temperature of the periocular, lacrimal caruncle, nasal, auricular, frontal-parietal, pelvic limb, torso, abdominal, lumbar, and thoracic regions was significantly higher during SJs (+3 °C) when compared to LJs (p < 0.0001). In particular, the frontal-parietal region had a significant increase of 10 °C during the post-transport phase (p < 0.0001) in both groups, recording the highest temperatures during this phase. Likewise, a strong positive significant correlation between the different regions was found (r = 0.90, p < 0.0001). It is worth mentioning that the herding, loading, pre-, and post-transport phases were the ones where the greatest thermal response was recorded, possibly due to the influence of human interaction. Finally, a strong positive correlation (r above 0.9, p > 0.001) between the periocular, lacrimal caruncle, pinna, and pelvic limb was found. According to the results, SJ could be considered a stressful event that hinders thermal generation, contrarily to LJ.

RESUMO

Finding tools to assess the stress response which can be easily applied, are non-invasive, reliable and measured in real time is still a relevant topic in many areas of biology. Vocal characteristics and temperature of certain body areas have been suggested to reflect HPA axis and ANS activation. We hypothesized that changes in vocalizations and peripheral body temperature will show the magnitude of the stress response, and that the change in these will covary. Our aim was to measure the change in vocal characteristics and eye and nasal temperature of kittens (n = 43 from nine litters of seven mixed-breed mothers) during a potentially stressful event and to test how these correlated. We found change in several vocal and thermal parameters during a short social separation. Our findings indicate that arousal due to ANS activation in kittens of the domestic cat resulted in an increasing number of vocalisations of longer duration and higher intensity, and in lower and a wider range in fundamental frequency. Calls also became less tonal with more jitter. Change in temperature was generally negative in the lacrimal caruncle as well as in the rhinarium, but with great variance across individuals. Change in eye temperature positively correlated with the intensity of the calls and the change in nose temperature positively correlated with the change in call length. The results suggest the continued difficulty in interpreting both physiological and behavioural data to assess an individual´s stress response.

Assuntos

RESUMO

Refinement is one of the principles aiming to promote welfare in research animals. The techniques used during an experimental protocol, including euthanasia selection, must prevent and minimize suffering. Although the current euthanasia methods applied to laboratory rodents are accepted, the controversial findings regarding the potential stress/distress they can cause is a field of research. The objective was to assess the thermal response of Wistar rats during various euthanasia methods using infrared thermography (IRT) to determine the method that prevents or diminishes the stress response and prolonged suffering. Pentobarbital (G1), CO2 (G2), decapitation (G3), isoflurane (G4), ketamine + xylazine (G5), and ketamine + CO2 (G6) were evaluated at five evaluation times with IRT to identify changes in the surface temperature of four anatomical regions: ocular (T°ocu), auricular (T°ear), interscapular (T°dor), and caudal (T°tai). Significant differences (p < 0.05) were found in G2 and G4, registering temperature increases from the administration of the drug to the cessation of respiratory rate and heart rate. Particularly, isoflurane showed a marked thermal response in T°ocu, T°ear, T°dor, and T°tai, suggesting that, in general, inhalant euthanasia methods induce stress in rats and that isoflurane might potentially cause distress, an effect that must be considered when deciding humane euthanasia methods in laboratory rodents.

RESUMO

Continuous ultrasound is recognized for its thermal effect and use in the tissue repair process. However, there is controversy about its dosage and efficacy. This study used infrared thermography, a non-invasive technique, to measure the short-term thermal effect of 3 MHz continuous ultrasound vs. a placebo, referencing the intensity applied. It was a single-blind, randomized clinical trial of 60 healthy volunteers (19-24 years old) divided into three equal groups. Group 1:1 W/cm2 for 5 min; Group 2: 0.5 W/cm2 for 10 min; and Group 3: the placebo for 5 min. The temperature was recorded through five thermographic images per patient: pre- and post-application, 5, 10, and 15 min later. After statistical analysis, a more significant decrease in temperature (p<0.05 ) was observed in the placebo group compared with the remaining groups after the application of continuous ultrasound. Group 1 was the one that generated the highest significant thermal effect (p<0.001), with an increase of 3.05 °C at 15 min, compared with the other two groups. It is concluded that to generate a thermal effect in the muscle, intensities of ≥1 W/cm2 are required, since the dosage maintained a temperature increase for more than 5 min.

RESUMO

Infrared thermography (IRT) is a technique used to diagnose Photovoltaic (PV) installations to detect sub-optimal conditions. The increase of PV installations in smart cities has generated the search for technology that improves the use of IRT, which requires irradiance conditions to be greater than 700 W/m2, making it impossible to use at times when irradiance goes under that value. This project presents an IoT platform working on artificial intelligence (AI) which automatically detects hot spots in PV modules by analyzing the temperature differentials between modules exposed to irradiances greater than 300 W/m2. For this purpose, two AI (Deep learning and machine learning) were trained and tested in a real PV installation where hot spots were induced. The system was able to detect hot spots with a sensitivity of 0.995 and an accuracy of 0.923 under dirty, short-circuited, and partially shaded conditions. This project differs from others because it proposes an alternative to facilitate the implementation of diagnostics with IRT and evaluates the real temperatures of PV modules, which represents a potential economic saving for PV installation managers and inspectors.

RESUMO

Thermography has grown in use in recent years. It is a valuable tool for measuring animal heat tolerance under heat stress conditions since it is a non-invasive, safe and practical methodology. Physiological variables such as respiration rate and eye temperature, and environmental variables such as air temperature and wet bulb temperature were analysed in animals from nine cattle breeds (Angus, Braford, Brangus, Canchim, Charolais, Hereford, Nelore, Simmental and Santa Gertrudis) and one bubaline (Mediterranean) at Rio Grande do Sul, Brazil. Positive correlations were observed between air temperature and respiration rate and eye temperature. Furthermore, the breed strongly influenced the eye temperature and respiration rate of the animals. Eye temperature showed strong correlation with air temperature and wet bulb temperature. Simmental and Nelore animals presented higher eye temperature values. Simmental presented alteration in respiratory rate before the other breeds and Nelore was the last breed to present this alteration. The inflection points in the broken line analysis indicated the environmental temperature limits at which breeds begin to change their respiration to compensate for environmental variation. The use of thermography has proven to be a technique with possible application to evaluate the temperature of animals. Logistic regression analysis allows us to observe how each breed behaves with the temperature change. Using respiration rates and eye temperatures it was possible to identify physiological limits for comfort in different breeds of bovine. In the future it would be interesting to conduct additional studies using other physiological variables and also other indices of climatic conditions.

Assuntos

RESUMO

Evaluating the welfare of buffaloes during transport is key to obtaining and commercializing high-quality meat products; however, effective assessments require recognizing several stressors that activate physiological mechanisms that can have repercussions on the health and productive performance of species. The aim of this study was to evaluate the surface temperatures of different body and head regions in this species during events prior, and posterior, to transport for short periods; that is, from paddock to loading. The second goal was to determine the level of correlation between thermal windows. This study used infrared thermography (IRT) to evaluate the surface temperature of 624 water buffaloes (Buffalypso breed) during 12 short trips (average duration = 2 h ± 20 min) by focusing on 11 regions of the body (Regio corporis), in the head regions (Regiones capitis) the face regions (Regiones faciei), Orbital region (Regio orbitalis) with special attention to structures such as the lacrimal caruncle, periocular area and lower eyelid (Regio palpebralis inferior); nasal region (Regio nasalis) with special attention to nostril thermal window; and regions of the skull (Regiones cranii) such as auricular region (Regio auricularis) with special attention to auditory canal and frontal-parietal region (Regio frontalis-parietalis) and trunk region (Truncus regionis) such as thoracic and abdominal regions, regions of the vertebral column (Columna vertebralis) with the thoracic vertebral region (Regio vertebralis thoracis) and lumbar region (Regio lumbalis); and regions of the pelvis limb (Regiones membri pelvini). Recordings were made during seven phases: paddock (P1), herding (P2), corral (P3), chute handling (P4), shipping (P5), pre- (P6), and post-transport (P7). A total of 48,048 readings were obtained from 11 thermal windows. The results showed that the surface temperatures of the windows increased by as much as 5°C during P2, P3, P5, P6, and P7 compared to P1 and P4 (p < 0.0001). Differences of at least 1°C were also observed between thermal windows in the craniofacial, lateral corporal, and peripheral zones (p < 0.0001). Finally, a strong positive correlation (r = 0.9, p < 0.0001) was found between the thermal windows. These findings lead to the conclusion that the surface temperature of the craniofacial and corporal regions of buffaloes transported for short periods varied in relation to the phase of mobilization (from paddock to post-transport), likely as a response to stressful factors, since herding and loading increased the thermal values in each window. The second conclusion is that there are strong positive correlations between central and peripheral thermal windows.

RESUMO

Landfill gas (LFG) is related to the biochemical processes generating heat and releasing CH4, CO2, and other gases in lower concentrations, which result in environmental impacts and risk of local explosion. Thermal infrared imagery (TIR) is employed to detect CH4 leakage as a risk control approach. However, the challenge for LFG leakage detection using TIR is establishing a relation between the gas flux and the ground temperature. This study evaluates the problem of a heated gas flowing through a porous medium column where the upward surface exchanges heat by radiation and convection to the environment. A heat transfer model that considers the upward LFG flow is proposed, and a sensibility analysis is developed to relate the flux to the ground temperature level in the condition of non-income solar radiation. An explicit equation to predict CH4 fugitive flow as a function of temperature anomalies of the ground was presented for the first time. The results show that the predicted ground surface temperatures are consistent with the literature's experimental observations. Moreover, the model was complementarily applied to a Brazilian landfill, with in situ TIR measurements in an area with a slightly fractured cover. In this field observation, the predicted CH4 flux was around 9025 g m-2 d-1. Model limitations concerning the soil homogeneity, the transient variation of atmospheric conditions or local pressure, and soil temperature difference in low-flux conditions (related to TIR-cameras accuracy) require further validation. Results could help landfill monitoring in conditions of a high-temperature ground anomaly in dry seasons.

Assuntos

RESUMO

Aim: In countries where access to mammography equipment and skilled personnel is limited, most breast cancer (BC) cases are detected in locally advanced stages. Infrared breast thermography is recognized as an adjunctive technique for the detection of BC due to its advantages such as safety (by not emitting ionizing radiation nor applying any stress to the breast), portability, and low cost. Improved by advanced computational analytics techniques, infrared thermography could be a valuable complementary screening technique to detect BC at early stages. In this work, an infrared-artificial intelligence (AI) software was developed and evaluated to help physicians to identify potential BC cases. Methods: Several AI algorithms were developed and evaluated, which were learned from a proprietary database of 2,700 patients, with BC cases that were confirmed through mammography, ultrasound, and biopsy. Following by evaluation of the algorithms, the best AI algorithm (infrared-AI software) was submitted to a clinic validation process in which its ability to detect BC was compared to mammography evaluations in a double-blind test. Results: The infrared-AI software demonstrated efficiency values of 94.87% sensitivity, 72.26% specificity, 30.08% positive predictive value (PPV), and 99.12% negative predictive value (NPV), whereas the reference mammography evaluation reached 100% sensitivity, 97.10% specificity, 81.25% PPV, and 100% NPV. Conclusions: The infrared-AI software here developed shows high BC sensitivity (94.87%) and high NPV (99.12%). Therefore, it is proposed as a complementary screening tool for BC.