TECRR: a benchmark dataset of radiological reports for BI-RADS classification with machine learning, deep learning, and large language model baselines.

BACKGROUND: Recently, machine learning (ML), deep learning (DL), and natural language processing (NLP) have provided promising results in the free-form radiological reports' classification in the respective medical domain. In order to classify radiological reports properly, a high-quality annotated and curated dataset is required. Currently, no publicly available breast imaging-based radiological dataset exists for the classification of Breast Imaging Reporting and Data System (BI-RADS) categories and breast density scores, as characterized by the American College of Radiology (ACR). To tackle this problem, we construct and annotate a breast imaging-based radiological reports dataset and its benchmark results. The dataset was originally in Spanish. Board-certified radiologists collected and annotated it according to the BI-RADS lexicon and categories at the Breast Radiology department, TecSalud Hospitals Monterrey, Mexico. Initially, it was translated into English language using Google Translate. Afterwards, it was preprocessed by removing duplicates and missing values. After preprocessing, the final dataset consists of 5046 unique reports from 5046 patients with an average age of 53 years and 100% women. Furthermore, we used word-level NLP-based embedding techniques, term frequency-inverse document frequency (TF-IDF) and word2vec to extract semantic and syntactic information. We also compared the performance of ML, DL and large language models (LLMs) classifiers for BI-RADS category classification. RESULTS: The final breast imaging-based radiological reports dataset contains 5046 unique reports. We compared K-Nearest Neighbour (KNN), Support Vector Machine (SVM), Naive Bayes (NB), Random Forest (RF), Adaptive Boosting (AdaBoost), Gradient-Boosting (GB), Extreme Gradient Boosting (XGB), Long Short-Term Memory (LSTM), Bidirectional Encoder Representations from Transformers (BERT) and Biomedical Generative Pre-trained Transformer (BioGPT) classifiers. It is observed that the BioGPT classifier with preprocessed data performed 6% better with a mean sensitivity of 0.60 (95% confidence interval (CI), 0.391-0.812) compared to the second best performing classifier BERT, which achieved mean sensitivity of 0.54 (95% CI, 0.477-0.607). CONCLUSION: In this work, we propose a curated and annotated benchmark dataset that can be used for BI-RADS and breast density category classification. We also provide baseline results of most ML, DL and LLMs models for BI-RADS classification that can be used as a starting point for future investigation. The main objective of this investigation is to provide a repository for the investigators who wish to enter the field to push the boundaries further.

Validation of the Mirai model for predicting breast cancer risk in Mexican women.

OBJECTIVES: To validate the performance of Mirai, a mammography-based deep learning model, in predicting breast cancer risk over a 1-5-year period in Mexican women. METHODS: This retrospective single-center study included mammograms in Mexican women who underwent screening mammography between January 2014 and December 2016. For women with consecutive mammograms during the study period, only the initial mammogram was included. Pathology and imaging follow-up served as the reference standard. Model performance in the entire dataset was evaluated, including the concordance index (C-Index) and area under the receiver operating characteristic curve (AUC). Mirai's performance in terms of AUC was also evaluated between mammography systems (Hologic versus IMS). Clinical utility was evaluated by determining a cutoff point for Mirai's continuous risk index based on identifying the top 10% of patients in the high-risk category. RESULTS: Of 3110 patients (median age 52.6 years ± 8.9), throughout the 5-year follow-up period, 3034 patients remained cancer-free, while 76 patients developed breast cancer. Mirai achieved a C-index of 0.63 (95% CI: 0.6-0.7) for the entire dataset. Mirai achieved a higher mean C-index in the Hologic subgroup (0.63 [95% CI: 0.5-0.7]) versus the IMS subgroup (0.55 [95% CI: 0.4-0.7]). With a Mirai index score > 0.029 (10% threshold) to identify high-risk individuals, the study revealed that individuals in the high-risk group had nearly three times the risk of developing breast cancer compared to those in the low-risk group. CONCLUSIONS: Mirai has a moderate performance in predicting future breast cancer among Mexican women. CRITICAL RELEVANCE STATEMENT: Prospective efforts should refine and apply the Mirai model, especially to minority populations and women aged between 30 and 40 years who are currently not targeted for routine screening. KEY POINTS: The applicability of AI models to non-White, minority populations remains understudied. The Mirai model is linked to future cancer events in Mexican women. Further research is needed to enhance model performance and establish usage guidelines.

Upper body thermal images and associated clinical data from a pilot cohort study of COVID-19.

OBJECTIVES: The data was collected for a cohort study to assess the capability of thermal videos in the detection of SARS-CoV-2. Using this data, a published study applied machine learning to analyze thermal image features for Covid-19 detection. DATA DESCRIPTION: The study recorded a set of measurements from 252 participants over 18 years of age requesting a SARS-CoV-2 PCR (polymerase chain reaction) test at the Hospital Zambrano-Hellion in Nuevo León, México. Data for PCR results, demographics, vital signs, food intake, activities and lifestyle factors, recently taken medications, respiratory and general symptoms, and a thermal video session where the volunteers performed a simple breath-hold in four different positions were collected. Vital signs recorded include axillary temperature, blood pressure, heart rate, and oxygen saturation. Each thermal video is split into 4 scenes, corresponding to front, back, left and right sides, and is available in MPEG-4 format to facilitate inclusion into pipelines for image processing. Raw JPEG images of the background between subjects are included to register variations in room temperatures.

3D print model for surgical planning in a case of recurrent osteoblastic osteosarcoma of the left maxilla. A case report.

Osteosarcoma (OS) of the head and neck is a rare and aggressive disease characterized by the formation of osteoid by malignant osteoblasts. The mandible or maxilla are the most common sites of presentation. Radiologically, these tumors show considerable, destructive growth with periosteal reaction, which can suggest the diagnosis of OS. 3D printing, as an emerging technology, can play a role in orthopedic oncology by providing patient-specific 3D printed models to improve surgical planning and facilitate patient understanding. We present the case of a male in his early 30s with a final histological diagnosis of recurrent osteosarcoma of the left maxilla, where a 3D printed model was helpful for the diagnostic workup, surgical planning, and the procedure.

Artificial Intelligence in Breast Imaging: A Special Focus on Advances in Digital Mammography & Digital Breast Tomosynthesis.

Breast cancer accounts for 30% of female cancers and is the second leading cause of cancerrelated deaths in women. The rate is rising at 0.4% per year. Early detection is crucial to improve treatment efficacy and overall survival of women diagnosed with breast cancer. Digital Mammography and Digital Breast Tomosynthesis have widely demonstrated their role as a screening tool. However, screening mammography is limited by radiologist's experience, unnecessarily high recalls, overdiagnosis, overtreatment and, in the case of Digital Breast Tomosynthesis, long reporting time. This is compounded by an increasing shortage of manpower and resources issue, especially among breast imaging specialists. Recent advances in image analysis with the use of artificial intelligence (AI) in breast imaging have the potential to overcome some of these needs and address the clinical challenges in cancer detection, assessment of treatment response, and monitoring disease progression. This article focuses on the most important clinical implication and future application of AI in the field of digital mammography and digital breast tomosynthesis, providing the readers with a comprehensive overview of AI impact in cancer detection, diagnosis, reduction of workload and breast cancer risk stratification.

NUSAP1 and PCLAF (KIA0101) Downregulation by Neoadjuvant Therapy is Associated with Better Therapeutic Outcomes and Survival in Breast Cancer.

Purpose: To evaluate whether changes in genomic expression that occur beginning with breast cancer (BC) diagnosis and through to tumor resection after neoadjuvant chemotherapy (NCT) reveal biomarkers that can help predict therapeutic response and survival. Materials and Methods: We determined gene expression profiles based on microarrays in tumor samples from 39 BC patients who showed pathologic complete response (pCR) or therapeutic failure (non-pCR) after NCT (cyclophosphamide-doxorubicin/epirubicin). Based on unsupervised clustering of gene expression, together with functional enrichment analyses of differentially expressed genes, we selected NUSAP1, PCLAF, MME, and DST. We evaluated the NCT response and the expression of these four genes in BC histologic subtypes. In addition, we study the presence of tumor-infiltrating lymphocytes. Finally, we analyze the correlation between NUSAP1 and PCLAF against disease-free survival (DFS) and overall survival (OS). Results: A signature of 43 differentially expressed genes discriminated pCR from non-pCR patients (|fold change >2|, false discovery rate <0.05) only in biopsies taken after surgery. Patients achieving pCR showed downregulation of NUSAP1 and PCLAF in tumor tissues and increased DFS and OS, while overexpression of these genes correlated with poor therapeutic response and OS. These genes are involved in the regulation of mitotic division. Conclusions: The downregulation of NUSAP1 and PCLAF after NCT is associated with the tumor response to chemotherapy and patient survival.

COVID-19 classification using thermal images.

SIGNIFICANCE: There is a scarcity of published research on the potential role of thermal imaging in the remote detection of respiratory issues due to coronavirus disease-19 (COVID-19). This is a comprehensive study that explores the potential of this imaging technology resulting from its convenient aspects that make it highly accessible: it is contactless, noninvasive, and devoid of harmful radiation effects, and it does not require a complicated installation process. AIM: We aim to investigate the role of thermal imaging, specifically thermal video, for the identification of SARS-CoV-2-infected people using infrared technology and to explore the role of breathing patterns in different parts of the thorax for the identification of possible COVID-19 infection. APPROACH: We used signal moment, signal texture, and shape moment features extracted from five different body regions of interest (whole upper body, chest, face, back, and side) of images obtained from thermal video clips in which optical flow and super-resolution were used. These features were classified into positive and negative COVID-19 using machine learning strategies. RESULTS: COVID-19 detection for male models [receiver operating characteristic (ROC) area under the ROC curve (AUC) = 0.605 95% confidence intervals (CI) 0.58 to 0.64] is more reliable than for female models (ROC AUC = 0.577 95% CI 0.55 to 0.61). Overall, thermal imaging is not very sensitive nor specific in detecting COVID-19; the metrics were below 60% except for the chest view from males. CONCLUSIONS: We conclude that, although it may be possible to remotely identify some individuals affected by COVID-19, at this time, the diagnostic performance of current methods for body thermal imaging is not good enough to be used as a mass screening tool.

Value of a breast imaging unit in the detection of breast cancer in Mexico.

The screening breast cancer detection rate in Mexico is low. The main objective of this study was to determine the breast cancer detection rate in a Mexican population that attended a breast imaging unit, in which the same radiologist comprehensively evaluated and interpreted breast imaging studies. A total of 5,429 mammograms performed between 2015 and 2016 were evaluated. Rates for biopsy indication, biopsies performed and positive biopsies for cancer were determined. The malignancy detection rate, after a comprehensive imaging evaluation in a breast imaging unit, was 24.3 per 1,000 mammograms. In symptomatic women was 52.9 per 1,000 mammograms, and in screening women was 11.1 per 1,000 mammograms. Breast imaging units in which a comprehensive imaging approach is performed represent an opportunity for low- and middle-income countries without population-based screening programs to achieve a more efficient detection of breast cancer, without generating a higher cost.

Spatial interaction between breast cancer and environmental pollution in the Monterrey Metropolitan Area.

This research examines the spatial structure of a sample of breast cancer (BC) cases and their spatial interaction with contaminated areas in the Monterrey Metropolitan Area (MMA). By applying spatial statistical techniques that treat the space as a continuum, degrees of spatial concentration were determined for the different study groups, highlighting their concentration pattern. The results indicate that 65 percent of the BC sample had exposure to more than 56 points of PM10. Likewise, spatial clusters of BC cases of up to 39 cases were identified within a radius of 3.5 km, interacting spatially with environmental contamination sources, particularly with refineries, food processing plants, cement, and metals. This study can serve as a platform for other clinical research by identifying geographic clusters that can help focus health policy efforts.

Subcutaneous trastuzumab with pertuzumab and docetaxel in HER2-positive metastatic breast cancer: Final analysis of MetaPHER, a phase IIIb single-arm safety study.

PURPOSE: Intravenous trastuzumab, pertuzumab, and docetaxel are first-line standard of care for patients with HER2-positive metastatic breast cancer (mBC). MetaPHER is the first study assessing the safety and tolerability of subcutaneous trastuzumab plus intravenous pertuzumab and chemotherapy in a global patient population with HER2-positive mBC. METHODS: In this open-label, single-arm, multicenter, phase 3b study, eligible patients were ≥ 18 years old with histologically/cytologically confirmed previously untreated HER2-positive mBC. All received ≥ 1 subcutaneous trastuzumab 600 mg fixed dose plus intravenous pertuzumab (loading dose: 840 mg/kg; maintenance: 420 mg/kg) and docetaxel (≥ 6 cycles; initial dose 75 mg/m2) every 3 weeks. The primary objective was safety and tolerability; secondary objectives included efficacy. RESULTS: At clinical cutoff, 276 patients had completed the study; median duration of follow-up was 27 months. The most common any-grade adverse events were diarrhea, alopecia, and asthenia; the most common grade ≥ 3 events were neutropenia, febrile neutropenia, and hypertension. There were no cardiac deaths and mean left ventricular ejection fraction was stable over time. Median investigator-assessed progression-free survival was 18.7 months; objective response rate was 75.6%. CONCLUSIONS: Safety and efficacy with subcutaneous trastuzumab plus intravenous pertuzumab and docetaxel in mBC are consistent with historical evidence of intravenous trastuzumab with this combination. Findings further support subcutaneous administration not affecting safety/efficacy profiles of trastuzumab in HER2-positive BC with increased flexibility in patient care. A fixed-dose combination of pertuzumab and trastuzumab for subcutaneous injection has recently been approved for the treatment of HER2-positive early/mBC, further addressing the increasing relevance of and need for patient-centric treatment strategies. TRIAL REGISTRATION: NCT02402712.

Impact of chemotherapy regimen and sequence on the effectiveness of scalp cooling for alopecia prevention.

PURPOSE: Scalp cooling (SC) is the most reliable method for the prevention of chemotherapy-induced alopecia. However, it remains unclear if its effectiveness is related to the chemotherapy regimen, sequence, and frequency. This study aims to evaluate SC performance among breast cancer patients who received different chemotherapy regimens. METHODS: The medical records of all consecutive patients undergoing curative-intent chemotherapy and receiving at least one SC session using the DigniCap® System from 2016-2020 in a private Mexican hospital were retrospectively reviewed. SC effectiveness according to chemotherapy regimen was analyzed using descriptive statistics. Successful alopecia prevention was defined as grade 0-1 alopecia (< 50% hair loss not requiring the use of a wig or headpiece) according to the Common Terminology Criteria for Adverse Events version 4.0. RESULTS: SC adequately prevented alopecia in 56/76 (74%) patients. All 12/12 (100%) and 15/17 (88%) patients receiving paclitaxel-only and docetaxel-based chemotherapy, respectively, had effective hair preservation. SC was successful in 7/16 (44%) patients when sequential chemotherapy started with anthracyclines and 22/30 (73%) when paclitaxel was administered upfront. Considering dose-dense regimens, 9/15 (60%) had satisfactory hair retention, and chemotherapy sequence was not clearly related to SC success. CONCLUSION: SC was highly effective in preventing alopecia, particularly with taxane-based regimens. Notably, better outcomes were observed when sequential chemotherapy started with taxanes followed by anthracyclines than when the inverse order was administered, suggesting that the chemotherapy sequence, rather than chemotherapeutic agents per se, might have a more significant impact on the effectiveness of SC for the prevention of alopecia.

Immunotyping of tumor-infiltrating lymphocytes in triple-negative breast cancer and genetic characterization.

Tumor-infiltrating lymphocytes (TILs) reflect the host immune response against cancer cells. Immunomodulators have been recently suggested as a novel therapeutic strategy against triple-negative breast cancer (TNBC). However, the TIL profile in TNBC has not been thoroughly investigated. In the present study, the percentage, immunophenotype and genetic profiles of TILs in pre-surgical tumor samples of patients with TNBC were evaluated prior to neoadjuvant chemotherapy (NAC). Patients diagnosed with breast cancer at Hospital San José TecSalud were consecutively and prospectively enrolled in the present study between August 2011 and August 2015. The pathological response to NAC was evaluated using the de Miller-Payne and MD Anderson Cancer Center system. TIL percentage (low, intermediate, and high) was evaluated using special hematoxylin-eosin staining on the core needle biopsies. The immunophenotype of TILs was assessed by immunohistochemistry (IHC) for CD3+, CD4+ and CD8+. In addition, the gene expression profile of CD3, CD4, CD8, CD20, CD45, forkhead box P3, interleukin 6, programmed cell death 1 and CD274 molecule was assessed in all patients. A total of 26 samples from patients with TNBC prior to NAC were included in the present study. TILs were low in 30.7%, intermediate in 38.4% and elevated in 30.7% of tumors. CD3+ and CD4+ counts were associated with the pathological response to NAC (P=0.04). Finally, an overexpression pattern of CD3, CD4, CD8, CD45 and CD20 genes was observed in patients with a partial or complete pathological response. The present results demonstrated that TILs may predict the pathological response to NAC in patients with TNBC. Furthermore, a more accurate association was identified between the high expression levels of CD3, CD4, CD8, CD45 and CD20 genes and partial and complete pathological response, compared with the association between high expression and IHC alone.

Comparison of characteristics in Mexican women with breast cancer according to healthcare coverage.

OBJECTIVE: To compare the sociodemographic, diagnostic, clinical, and treatment-related characteristics and outcomes of patients with breast cancer in two hospitals in Mexico according to type of healthcare coverage. METHODS: A retrospective cohort study of women with breast cancer according to public or private healthcare coverage in two hospitals was done. Patients were treated by the same group of physicians and healthcare infrastructure. Groups were compared using the chi-square test for categorical variables, Mann-Whitney U-test and Student's t-test for quantitative variables, and Kaplan-Meier estimator and log-rank test for time dependent outcomes (including recurrence-free and overall survival). A value of p < 0.05 was considered statistically significant. RESULTS: A total of 282 women were included. Mean age at diagnosis was 52 years. Women with public healthcare coverage were diagnosed more frequently with self-detected tumors (82.8% vs 47.9%, p < 0.001) and advanced clinical stage (III and IV) (31.1% vs 17.8%, p = 0.014). More women with public healthcare insurance underwent initial systemic treatment (41.1% vs 17.8%, p < 0.001) and mastectomy (70.1% vs 54.9%, p = 0.020), and received more chemotherapy (79.4% vs 43.8%, p < 0.001) and adjuvant radiotherapy (68.9% vs 53.4%, p = 0.017). Overall, no differences were found in survival outcomes according to healthcare coverage. Trends suggesting worse recurrence-free and overall survival were observed in patients with public coverage at 3 years follow-up in stage III (85.7% vs 67.3% and 100% vs 84.6%, respectively) and triple negative disease (83.3% vs 74.5% and 100% vs 74.1%, respectively). CONCLUSION: Strategies to promote preventive medicine, diagnostic mammograms, and prompt diagnosis of breast cancer in Mexican women with public health coverage are needed. Access to the main treatment modalities by Seguro Popular and good quality care by an experienced group of physicians likely explains the similar outcomes between patients with private and public healthcare coverage. However, trends suggesting worse survival for patients with public medical coverage with stage III and triple-negative disease should encourage close follow-up.

Does axillary lymph node size predict better metastatic involvement than apparent diffusion coefficient (ADC) value in women with newly diagnosed breast cancer?

BACKGROUND: It has been demonstrated that the number of metastatic axillary lymph nodes (mALNs) influence disease-free and overall survival in patients with breast cancer. PURPOSE: To determine if the ALN size is more accurate than the ALN apparent diffusion coefficient (ADC) value to predict metastatic involvement in newly diagnosed breast cancer. MATERIAL AND METHODS: A total of 44 patients with breast cancer were included. Magnetic resonance imaging (MRI) examinations were performed on a 1.5-T system with sagittal T1-weighted fast spin-echo non-fat saturated, sagittal T2-weighted fast spin-echo non-fat saturated, axial diffusion-weighted imaging echo-planar (b values of 0 and 700 s/mm2), and non-contrast axial VIBRANT sequences. The size and the ADC value were obtained for ALN ipsilateral and contralateral to breast cancer. The reference standard was the histopathologic lymph node status. RESULTS: mALN had a greater cortical thickness compared to contralateral non-mALN (10.3 ± 5.32 mm vs. 4 ± 1.17 mm, P ≤ 0.001). The threshold of ≥6.7 mm for predicting axillary metastatic involvement had a sensitivity and a specificity of 80.0% and 97.7%, respectively. The ADC value of mALN was significantly higher than the contralateral non-mALN (0.90 ± 0.12 × 10-3mm2/s vs. 0.78 ± 0.12 × 10-3mm2/s; P = 0.001). The threshold of ≥0.86 × 10-3mm2/s had a sensitivity and a specificity of 66.7% and 76.7%, respectively. CONCLUSION: Our results indicate that the cortical thickness has a better diagnostic performance in the differentiation of metastatic and non-metastatic lymph nodes than the lymph node ADC.

Comparison of Automated and Manual DNA Isolation Methods of Liquid-Based Cytology Samples.

Liquid-based cytology (LBC) has been used as a diagnostic tool for cervical cancer for years and is now being adopted for other gynecological cancers. LBC represents an important challenge to ensure that the process yields representative biospecimens for quality control (QC) of diagnostic procedures. In this study, we compare QC parameters (integrity, yield and purity, and polymerase chain reaction [PCR] amplification) of DNA isolated from LBC (N = 296) using two different nucleic acid isolation methods, manual (n = 233) or automated (n = 63). We also evaluated two different types of cytological brushes for sampling from the cervix. Our results suggest that manual isolation (yield 22.81 ± 1.92 µg) resulted in increased DNA recovery when compared with automated isolation (yield 9.96 ± 1.11 µg) from LBC samples, with a p-value of <0.0003. We estimated that 98% (53/54) of the samples preserved the integrity of DNA and were suitable for standard molecular biology analyses. The ß-globin gene was amplified in 100% (296/296) of the DNA samples by endpoint PCR. We found no significant difference between the performance of the cytological brushes (p value of <0.6711) in a general overview. However, when looking at the results from using each brush individually, the manual isolation method was statistically superior to the automated method. Our work illustrates the impact of good QC of preanalytic conditions, which will be important for the application of LBC for developing early detection methods for gynecological cancers.

Genetic alterations of triple negative breast cancer (TNBC) in women from Northeastern Mexico.

Triple negative breast cancer (TNBC) is a subtype of breast cancer of heterogeneous nature that is negative for estrogen receptor (ER), progesterone receptor (PR) and growth factor human epidermal 2 (HER2) following immunohistochemical analysis. TNBC is frequently characterized by relapse and reduced survival. To date, there is no targeted therapy for this type of cancer. Chemotherapy, radiotherapy, and surgery remain as the standard treatments options. The lack of a target therapy and the heterogeneity of TNBC highlight the need to seek new therapeutic options. In this study, fresh tissue samples of TNBC were analyzed with a panel of 48 driver genes (212 amplicons) that are likely to be therapeutic targets. We found intron variants, missense, stop gained and splicing variants in TP53, PIK3CA and FLT3 genes. Interestingly, all the analyzed samples had at least two variants in the TP53 gene, one being a drug response variant, rs1042522, found in 94% of our samples. We also found seven additional variants not previously reported in the TP53 gene, to the best of our knowledge, with probable deleterious characteristics of the tumor suppressor gene. We found four genetic variants in the PIK3CA gene, including two missense variants. The rs2491231 variant in the FLT3 gene was identified in 84% (16/19) of the samples, which not yet reported for TNBC, to the best of our knowledge. In conclusion, genetic variants in TP53 were found in all TNBC tumors, with rs1042522 being the most frequent (94% of TNBC biopsies), which had not been previously reported in TNBC. Also, we found two missense variants in the PIK3CA gene. These results justify the validation of these genetic variants in a large cohort, as well as the extensive study of their impact on the prognosis and therapy management of TBNC.

A solitary presentation of panniculitis in a patient with a history of breast cancer.

INTRODUCTION: Panniculits presents as an inflammation of the subcutaneous adipose tissue of the skin. In breast, panniculitis is very rare and is usually a manifestation of underlying inflammatory conditions. The typical presentation is palpable tender nodules, which in cases of breast panniculitis, triggers an extensive work up to exclude a malignancy. Herein we present a case of septal and lobar panniculitis in a female with clinical history of invasive ductal carcinoma. PRESENTATION OF THE CASE: A 52-year old female with past medical history of invasive breast carcinoma 5 years prior to the presentation. The patient's chief complaint was a 1-year history of a subcutaneous nodular lesion on her left breast. A core biopsy of the firm nodule showed marked inflammation of the breast. A second skin biopsy showed an abundant chronic inflammatory infiltrate, with lymphocytic vasculitis and neuritis, suggestive of an underlying autoimmune process. DISCUSSION: Subcutaneous panniculitis with or without vasculitis is a rare condition when presenting in the breast. Panniculitis can mimic malignancy and thus, it is important to differentially diagnose it from breast carcinoma. Histologically, it is classified in lobular and septal lymphocytic panniculitis depending on specific diagnostic characteristics. CONCLUSION: Panniculitis of the breast is a rare condition that needs to be included in the differential diagnosis of subcutaneous breast masses. In all cases, but specifically in females with history of breast cancer, panniculitis still should be thought of as a possibility, and imaging as well as other diagnostic techniques can aid in making the correct diagnosis.

Pure mucinous breast carcinoma in a 25-year-old female, a case report.

INTRODUCTION: Mucinous carcinoma is a variant of invasive breast carcinomas representing 2% of them. These tumors frequently develop in postmenopausal females; it is a rare histological variant in young patients. CASE PRESENTATION: A 25-year-old female refers a slow growth mass of 2 years of evolution. Excisional biopsy reveals a pure mucinous carcinoma with positive hormone receptors and negative HER2. She was treated with hormone therapy and surgical resection. DISCUSSION: Mucinous carcinoma is a rare variant reported in young patients. Many series report that is frequently found in postmenopausal patients. We present a case of a pure mucinous carcinoma in a 25-year-old female with the importance of being a low-frequency malignancy in young patients. CONCLUSION: Due to its benign course, it is important to know that this lesion can also present in young patients. The importance underlies in the multidisciplinary management at the right time in a proper way.

Radiogenomics analysis identifies correlations of digital mammography with clinical molecular signatures in breast cancer.

In breast cancer, well-known gene expression subtypes have been related to a specific clinical outcome. However, their impact on the breast tissue phenotype has been poorly studied. Here, we investigate the association of imaging data of tumors to gene expression signatures from 71 patients with breast cancer that underwent pre-treatment digital mammograms and tumor biopsies. From digital mammograms, a semi-automated radiogenomics analysis generated 1,078 features describing the shape, signal distribution, and texture of tumors along their contralateral image used as control. From tumor biopsy, we estimated the OncotypeDX and PAM50 recurrence scores using gene expression microarrays. Then, we used multivariate analysis under stringent cross-validation to train models predicting recurrence scores. Few univariate features reached Spearman correlation coefficients above 0.4. Nevertheless, multivariate analysis yielded significantly correlated models for both signatures (correlation of OncotypeDX = 0.49 ± 0.07 and PAM50 = 0.32 ± 0.10 in stringent cross-validation and OncotypeDX = 0.83 and PAM50 = 0.78 for a unique model). Equivalent models trained from the unaffected contralateral breast were not correlated suggesting that the image signatures were tumor-specific and that overfitting was not a considerable issue. We also noted that models were improved by combining clinical information (triple negative status and progesterone receptor). The models used mostly wavelets and fractal features suggesting their importance to capture tumor information. Our results suggest that molecular-based recurrence risk and breast cancer subtypes have observable radiographic phenotypes. To our knowledge, this is the first study associating mammographic information to gene expression recurrence signatures.