Breast cancer comprises a serious public health concern. The three primary techniques for detecting breast cancer are ultrasound, mammography, and magnetic resonance imaging (MRI). However, the existing methods of diagnosis are not practical for regular mass screening at short time intervals. Thermography could be a solution to this issue because it is a non-invasive and low-cost method that can be used routinely as a self-screening method. The research significance of this work lies in the implementation and integration of multiple different AI techniques for achieving diagnosis based on breast thermograms from several data sources. The data sources contain 306 images. The concept of transfer learning with several pre-trained models is implemented. Bayesian Networks (BNs) are also used to have interpretability of the diagnosis. A novel feature extraction from images (related to temperature) has been implemented and feeds the BNs. Finally, all methods and the classification results of pre-trained models are compared. It is found that the best result amongst the transfer learning concept is achieved with MobileNet, which delivered 93.8% accuracy. Furthermore, the BN achieves an accuracy of 90.20%, and finally, the expert model that combines CNNs and BNs gives an accuracy of 90.85%, even with a limited amount of data available. The integration of CNN and BN aims to overcome the hardship of interpretability. These approaches demonstrate high performance with added interpretability compared to previous works. In conclusion, the deep neural network provides promising results in breast cancer detection. It could be an ideal candidate for Breast Self-Exam (BSE), the goal recommended by WHO for mass screening.

Breast cancer remains one of the most prevalent cancers worldwide, necessitating reliable, efficient, and precise diagnostic methods. Meanwhile, the rapid development of artificial intelligence (AI) presents significant opportunities for integration into various fields, including healthcare, by enabling the processing of medical data and the early detection of cancer. This review examines the major medical imaging techniques used for breast cancer detection, specifically mammography, ultrasound, and thermography, and identifies widely used publicly available datasets in this domain. It also surveys traditional machine learning and deep learning approaches commonly applied to the analysis of mammographic, ultrasound, and thermographic images, discussing key studies in the field and evaluating the potential of different AI techniques for breast cancer detection. Furthermore, the review highlights the development and integration of explainable artificial intelligence (XAI) to enhance transparency and trust in medical imaging-based diagnoses. Finally, it considers potential future directions, including the application of large language models (LLMs) and multimodal LLMs in breast cancer diagnosis, emphasizing recent research aimed at advancing the precision, accessibility, and reliability of diagnostic systems.

Background: Mammography is the preferred method for the diagnosis of breast cancer. However, this diagnostic technique fails to detect tumors of small sizes, and it does not work well for younger patients with high breast tissue density. Methods: This paper proposes a novel tool for the early detection of breast cancer, which is patient-specific, non-invasive, inexpensive, and has potential in terms of accuracy compared with existing techniques. The main principle of this method is based on the use of temperature contours from breast skin surfaces through thermography, and inverse thermal modeling based on Finite Element Analysis (FEA) and a Genetic Algorithm (GA)-based optimization tool to estimate the depths and sizes of tumors as well as patient/breast-specific tissue properties. Results: The study was conducted by using a 3D geometry of patients’ breasts and their temperature contours, which were clinically collected using a 3D scanner and a thermal imaging infrared (IR) camera. Conclusion: The results showed that the combination of 3D breast geometries, thermal images, and inverse thermal modeling is capable of estimating patient/breast-specific breast tissue and physiological properties such as gland and fat contents, tissue density, thermal conductivity, specific heat, and blood perfusion rate, based on a multilayer model consisting of gland and fat. Moreover, this tool was able to calculate the depth and size of the tumor, which was validated by the doctor’s diagnosis.

Background. Post-intubation cicatricial laryngeal stenosis (PICLS) represents one of the most severe long-term complications of pediatric airway management. By systematically analyzing clinical and procedural variables across different grades of PICLS, this study addresses a critical gap in pediatric airway research and provides clinically relevant descriptive data on stenosis severity. Materials and methods. A retrospective single-center case-series study was conducted and included pediatric patients (0–18 years) treated for PICLS at a tertiary referral pediatric otolaryngology center between 2016 and 2024. Spearman correlation and multiple regression analyses were used to evaluate possible associations between clinical factors and stenosis grade. Results. Among 172 children with PICLS, severe forms of stenosis (Grades 3–4) were observed in 37.2%, with predominant subglottic localization (85.3%). Age at primary intubation (p = 0.02) and the type of intubation (emergency/elective; p = 0.04) were the only variables significantly associated with stenosis severity in this cohort, whereas sex, reintubation, comorbidities, and delivery-related factors showed no significant associations. Mild stenosis (Grades 1–2) more frequently followed intubation for elective surgery and infections, whereas severe stenosis was more commonly associated with intubation due to central nervous system pathology and infections. Conclusions. Age at primary intubation and the type of intubation (emergency/elective) were associated with stenosis severity in this cohort. These findings should be interpreted in light of the retrospective case-series design and the absence of a control group, but they may contribute to improved clinical characterization of PICLS severity in children.

This paper presents a three-dimensional numerical study of temperature patterns in a realistic multi-layered model of a female breast including blood perfusion. The breast surface temperature distributions are computed and analyzed with different tumor positions, sizes and different fat contents in the breast. The results are compared with experimental results for validation of the model. The paper shows that realistic breast model can accurately predict the temperature distributions inside the breast compared with traditional idealized models. The results demonstrate that all of the identifiable tumor occurrences were at the depth from 13 mm to 23 mm while none of the tumors at a depth of 29 mm were found to be detected. In respect to this, it was observed that tumors lied in the gland layer had less impact on the temperature profile of the breast. In addition, it was perceived that because of the natural deformation the breast geometry has an asymmetric surface temperature distribution in regards to symmetric tumor positions. Thus, the conducted parametric study analyzes the tumor location, size, and metabolic heat generation, and compares different temperature patterns subjected to the changes in the fat layer. Additionally, this study uses more realistic breast geometry model compared to previous studies. All this gives greater insight into the detectability of tumors with a variety of physiological conditions based on personalized patients’ data and can give useful insight to improve the accuracy of computer-aided diagnosis using similar breast models. This can provide a very useful tool in inverse thermal modelling for the accurate detection of tumors in the breast.

Ne of the most common causes of disease related death among young women in almost every country in the world is the breast cancer. Valid and timely diagnosis of the breast cancer is vital, as its earlier identification considerably helps any further treatment. There are several methods for breast cancer identification. This paper acknowledges that the gold standard method of breast cancer identification is mammography, which can be further assisted with the adjunctive tool of thermography. For both these techniques there are many research approaches that use computer aided detection systems to improve the detectability of the breast cancer which are based on these basic methods. The developments are mainly based on the recent progress in the field of machine learning techniques, numerical simulation, and statistical methods. They span a broad range of the artificial intelligence (AI) field. The paper elaborates on these recent and future paths of progress in the field of artificial intelligence for breast cancer...