Deep Learning in Breast Cancer Imaging: State of the Art and Recent Advancements in Early 2024.

The rapid advancement of artificial intelligence (AI) has significantly impacted various aspects of healthcare, particularly in the medical imaging field. This review focuses on recent developments in the application of deep learning (DL) techniques to breast cancer imaging. DL models, a subset of AI algorithms inspired by human brain architecture, have demonstrated remarkable success in analyzing complex medical images, enhancing diagnostic precision, and streamlining workflows. DL models have been applied to breast cancer diagnosis via mammography, ultrasonography, and magnetic resonance imaging. Furthermore, DL-based radiomic approaches may play a role in breast cancer risk assessment, prognosis prediction, and therapeutic response monitoring. Nevertheless, several challenges have limited the widespread adoption of AI techniques in clinical practice, emphasizing the importance of rigorous validation, interpretability, and technical considerations when implementing DL solutions. By examining fundamental concepts in DL techniques applied to medical imaging and synthesizing the latest advancements and trends, this narrative review aims to provide valuable and up-to-date insights for radiologists seeking to harness the power of AI in breast cancer care.

Locoregional staging of breast cancer: contrast-enhanced mammography versus breast magnetic resonance imaging.

PURPOSE: Breast cancer diagnosis often involves assessing the locoregional spread of the disease through MRI, as multicentricity, multifocality and/or bilaterality are increasingly common. Contrast-enhanced mammography (CEM) is emerging as a potential alternative method. This study compares the performance of CEM and MRI in preoperative staging of women with confirmed breast carcinoma. Patients were also asked to fill in a satisfaction questionnaire to rate their comfort level with each investigation. METHODS: From May 1st, 2021 to May 1st, 2022, we enrolled 70 women with confirmed breast carcinoma who were candidates for surgery. For pre-operative locoregional staging, all patients underwent CEM and MRI examination, which two radiologists evaluated blindly. We further investigated all suspicious locations for disease spread, identified by both CEM and MRI, with a second-look ultrasound (US) and eventual histological examination. RESULTS: In our study cohort, MRI and CEM identified 114 and 102 areas of focal contrast enhancement, respectively. A true discrepancy between MRI and CEM occurred in 9 out of 70 patients examined. MRI reported 8 additional lesions that proved to be false positives on second-look US in 6 patients, while it identified 4 lesions that were not detected by CEM and were pathological (true positives) in 3 patients. CONCLUSIONS: CEM showed results comparable to MRI in the staging of breast cancer in our study population, with a high rate of patient acceptability.

Optimisation of protection in the medical exposure of recurrent adult patients due to computed tomography procedures: development of recurrent exposures reference levels.

OBJECTIVES: To assess the incidence (1 year) and the cumulative incidence (3 years) of the condition of patients accruing cumulative effective doses (CED) of ≥ 100 mSv and their variability among different hospitals. To establish and validate a reference level for the CED in patients with recurrent exposures (RERL) and provide a RERL value. METHODS: Data of CT exposure was collected in 9 similar hospitals. The database included 294,222 patient*years who underwent 442,278 CT exams in 3 years. The incidence proportion of patients with CED ≥ 100 mSv in a given year (I100;1) and the 3-year cumulative incidence of patients with CED ≥ 100 mSv over 3 consecutive years (I100;3) were calculated and compared among different institutions. RESULTS: I100;1 ranged from a minimum of 0.1% to a maximum of 5.1%. The percentage of recurrent patients was quite uniform among centres ranging from 23 to 38%. The I100;3 ranged from a minimum of 1.1 to 11.4%. There was a strong positive correlation between the third quartile values of yearly CED and yearly incidence (r = 0.90; R2 = 0.81; p < 0.0001). RERL value in our study was found at 34.0 mSv. CONCLUSION: The management of patients with recurrent exposures is highly variable among hospitals leading to a 50-fold variation in I100;1 and to a tenfold variation in I100;3. RERL could be established and used by taking as a RERL quantity the CED and as a RERL value the 75th percentile of the third quartiles of the distribution of the yearly CED obtained by surveying different hospitals. CLINICAL RELEVANCE STATEMENT: This is the first ever multicentre study that quantifies recurrent exposures in terms of incidence and cumulative incidence of patients with CED ≥ 100 mSv. RERL establishment and use could benefit the optimisation of radioprotection of patients with recurrent exposures. KEY POINTS: This is the first multicentre study estimating yearly incidence and 3-year cumulative incidence of patients with cumulative effective doses ≥ 100 mSv. In this study, a 50-fold inter centre variation between the maximum (5.1%) and the minimum value (0.1%) of yearly incidence of patients with cumulative effective doses ≥ 100 mSv was reported. The range of the 3-year cumulative incidence extended from 1.1 to 11.4% (a tenfold variation) The third quartile of the yearly cumulative effective doses in a centre showed a strong positive correlation with the yearly incidence of patients with cumulative effective doses ≥ 100 mSv, with a potential of being used to set reference levels for recurrent exposures.

Early patello-femoral condropathy assessment through quantitative analyses via T2 mapping magnetic resonance after anterior cruciate ligament reconstruction.

BACKGROUND: Patellar femoral chondropathy (FPC) is a common problem in patients undergoing anterior cruciate ligament reconstruction (ACL-R) surgery, which, if left untreated, predisposes to arthrosis. Magnetic resonance imaging (MRI) is the non-invasive gold standard for morphological evaluation of cartilage, while in recent years advanced MRI techniques (such as T2 mapping) have been developed to detect early cartilage biochemical changes. This study evaluates the different onset of early PFC between B-TP-B and HT through T2 mapping. Secondly, it aims to assess the presence of any concordance between self-reported questionnaires and qualitative MRI. MATERIALS AND METHODS: 19 patients enrolled were divided into two groups based on the type of intervention: B-PT-B and HT. After a median time of 54 months from surgery, patients were subjected to conventional MRI, T2 mapping, and clinical-functional evaluation through three self-reported questionnaires: Knee Injury and Osteoarthritis index (KOOS); Tegner Lysholm Knee Scoring Scale; International Knee Documentation Committee (IKDC). RESULTS: There is not statistically significant difference in the comparison between the two MRI techniques and the two reconstructive techniques. KOOS and Tegner Lysholm scales showed significant agreement with MRI results on the grading of chondropathy. CONCLUSIONS: There are no differences between B-TP-B and HT techniques in the early development of PFC detectable through non-invasive methods. Due to the large reduction in the frequency of physical activity following ACL-R and the finding of mild PFC (grade I and II) in a substantial proportion of patients, after a relatively short period from ACL-R, all patients should undergo conservative treatment.

Capecitabine-Induced Ileitis during Neoadjuvant Pelvic Radio-Chemotherapy for Locally Advanced Rectal Cancer: A Case Report with Literature Review.

We report on a clinical case of capecitabine-induced acute ileitis in a patient treated with pre-operative concurrent chemoradiation with capecitabine for locally advanced rectal cancer and provide a comprehensive literature review. This a rare, but life-threatening, clinical situation, that clinicians should be aware of. Severe persistent diarrhea is the most frequent clinical feature and computed tomography is a valid tool for diagnosis. Conservative management includes capecitabine withdrawal, antidiarrheal therapy and endovenous hydration, together with dietary modifications and broad-spectrum antibiotics. Pelvic irradiation represents an adjunctive risk factor, which may increase the likelihood of occurrence of terminal ileitis. Early recognition and prompt intervention are crucial for successful clinical management.

Exploring the potential of contrast agents in breast cancer echography: current state and future directions.

Breast cancer stands as the most frequent malignancy and leading cause of death among women. Early and accurate detection of this pathology represents a crucial factor in enhancing both incidence and mortality rates. Ultrasound (US) examination has been extensively adopted in clinical practice due to its non-invasiveness, affordability, ease of implementation, and wide accessibility, thus representing a valuable first-line diagnostic tool for the study of the mammary gland. In this scenario, recent developments in nanomedicine are paving the way for new interpretations and applications of US diagnostics, which are becoming increasingly personalized based on the molecular phenotype of each tumor, allowing for more precise and accurate evaluations. This review highlights the current state-of-the-art of US diagnosis of breast cancer, as well as the recent advancements related to the application of US contrast agents to the field of molecular diagnostics, still under preclinical study.

Mammary nodules as incidental findings on chest computed tomography: a retrospective analysis on their frequency and predictive value.

PURPOSE: To evaluate the frequency of mammary nodules as incidental findings on chest CT scans and to determine a correlation between semiological features and mammographic and histopathological outcomes. METHODS: A total of 42,864 chest CT scans performed on patients with breast-unrelated working diagnoses by the Radiology Department at AOU Maggiore della Carità, between 1st January 2016 and 30th April 2022, were analysed. Sixty-eight patients (3 males and 65 females) with mammary nodule CT detection were selected and subjected to mammography, mammary ultrasound and, eventually, biopsy. RESULTS: Thirty-five of the 68 patients received a histopathological confirmation of malignancy. According to Pearson's Chi-square test, the CT features most likely associated with BI-RADS 5 following mammography were post-contrast enhancement (p = 0.001), margin irregularity (p = 0.0001), nipple retraction (p = 0.001), skin thickening (p = 0.024), and the presence of structurally atypical lymph nodes suspicious for metastatic involvement (p = 0.0001). The CT features predictive of a biopsy positive for malignancy were post-contrast enhancement (p = 0.0001), margin irregularity (p = 0.0001), and the presence of suspicious lymph nodes (p = 0.011). Finally, 63.4% of patients with a working diagnosis related to cancer were diagnosed with breast cancer. CONCLUSION: Chest CT incidental findings of mammary nodules had a 0.21% incidence rate. The accurate description of some CT scan features, such as post-contrast enhancement, margin irregularity, nipple retraction, skin thickening and the presence of structurally atypical lymph nodes, may help to establish a radiological suspicion of malignancy, especially if these characteristics are supported by a working diagnosis of cancer.

Cumulative radiation exposure from radiological imaging in patients with Hodgkin and diffuse large b-cell lymphoma not submitted to radiotherapy.

OBJECTIVE: To assess the cumulated exposure to radiation due to imaging in Hodgkin (HL) and diffuse large B-cell (DLBCL) lymphoma patients who were not submitted to radiotherapy. METHODS: The study population included 51 and 83 adult patients with HL and DLBCL, with a follow-up duration >1 year. The cumulated exposure was expressed using patient-specific data as cumulated effective dose (CED). RESULTS: Fifty-one HL patients (median age 47 years) were followed for a median of 3.5 years. The median total CED per subject was 104 mSv. CT and PET/CT examinations accounted for 75 and 25% of the total CED, respectively. 26 patients (49%) had a total CED ≥ 100 mSv and the maximum CED was 302 mSv. Eighty-three DLBCL patients (median age 66 years) were followed for a median of 3.7 years. The median total CED per subject over the study period was 134 mSv. CT and PET/CT for 86% and 13% of the total CED, respectively. 56 patients (67%) had a total CED ≥100 mSv. The maximum CED was 557 mSv. CONCLUSION: Our study demonstrated the large number of imaging procedures performed for patients with lymphoma. Overall, 61% of the patients accrued a CED ≥ 100 mSv. Imaging policies were only in a partial agreement with current international guidelines. ADVANCES IN KNOWLEDGE: The cumulated exposure radiation exposure may be of concern in HL patients and the contribution of CT procedures to the total CED is significant. The standardisation of clinical guidelines for managing patients with lymphoma is warranted.

Vascular Complications following Vacuum-Assisted Breast Biopsy (VABB): A Case Report and Review of the Literature.

INTRODUCTION: Vacuum-assisted breast biopsy (VABB) has been evaluated as a minimally invasive, safe, and accurate procedure with low complication risks; the most frequent one is the mild/moderate hematoma, which occurs with a low-frequency rate, and the majority of patients who experienced it can be treated successfully with only manual compression and dressing. Although cases of uncontrollable catastrophic bleeding are exceedingly rare, local breast vessel involvement is a concrete risk, even in patients with no bleeding propensity. CASE PRESENTATION: In this article, we aimed to describe a 60 years-old woman who, following VABB, experienced a massive hematoma without external bleeding and was successfully treated with embolization. The woman was called back for a cluster of suspicious microcalcifications identified in the left breast's upper-outer quadrant; however, following histopathological analysis, the few samples collected were negative. She had a silent past medical history, 100% performance status, and no active pharmacotherapy. Approximately 15-30 min after VABB, the patient complained of weakness, pain, and lipothymia. A physical examination revealed a massive hematoma without external bleeding. Clinical data reported PaO 65/40 mmHg and blood chemistry Hb < 10 g/dL. The emergency team was alerted to stabilize the patient, and after that, the breast hemorrhage was controlled by endovascular embolization. Despite this being a rare occurrence, it is important to draw up and follow an appropriate protocol to ensure proper patient management and early treatment. DISCUSSION: This case illustrates the prompt and accurate management of a rare complication following VABB. Due to the very high number of patients undergoing this particular procedure, we aim to point out the concrete risk of vascular injury; other similar cases are described to support our thesis and provide different clinical manifestations of this rare occurrence.

Innovation in Acute Ischemic Stroke Patients over 80 y/o-A Retrospective Monocentric Study on Mechanical Thrombectomy of Consecutive Patients: Is Age an Adequate Selection Criterion?

BACKGROUND: Although it is clear that stroke is a time-dependent and age-associated disease, we still need more evidence regarding the efficacy and outcomes in elderly patients who were excluded from the first trials of mechanical thrombectomy. The aim of this study is to highlight patient characteristics, the timing of medical attention and therapy, successful recanalization, and functional outcomes in patients over 80 y/o who underwent mechanical thrombectomy at the Ospedale Maggiore della Carità di Novara (Hub) since endovascular stroke treatment was first started here. METHODS: all 122 consecutive patients over 80 y/o at admission who underwent mechanical thrombectomy between 2017 and 2022 at our Hub center were retrospectively included in our database. A good functional outcome in these elderly patients was considered as the 90 days modified Rankin Scale (mRS) ≤ 3 and/or a decrease in functional status as ∆mRS ≤ 1 in order to interpret the results for patients with intact intellect and basal mRS > 3. Successful recanalization as a score of TICI ≥ 2b (Thrombolysis in Cerebral Infarction) was analyzed as a secondary outcome. RESULTS: Good functional outcome (mRS ≤ 3 and/or ∆mRS ≤ 1) was observed in 45.90% (56/122). The rate of successful recanalization (TICI ≥ 2b) was 65.57% (80/122). CONCLUSION: Our data confirm that a good outcome in the elderly age group has a correlation with age; being younger, with a milder NIHSS (National Institutes of Health Stroke Scale) at the onset and with a lower pre-morbid mRS is statistically associated with a better outcome. However, age should not be a criterion to exclude older patients from mechanical thrombectomy. Decision-making should take into consideration the pre-morbid mRS and the severity of the stroke on the NIHSS scale, especially in the age group over 85 y/o.

The Non-Invasive Diagnosis of Chronic Coronary Syndrome: A Focus on Stress Computed Tomography Perfusion and Stress Cardiac Magnetic Resonance.

Coronary artery disease is still a major cause of death and morbidity worldwide. In the setting of chronic coronary disease, demonstration of inducible ischemia is mandatory to address treatment. Consequently, scientific and technological efforts were made in response to the request for non-invasive diagnostic tools with better sensitivity and specificity. To date, clinicians have at their disposal a wide range of stress-imaging techniques. Among others, stress cardiac magnetic resonance (S-CMR) and computed tomography perfusion (CTP) techniques both demonstrated their diagnostic efficacy and prognostic value in clinical trials when compared to other non-invasive ischemia-assessing techniques and invasive fractional flow reserve measurement techniques. Standardized protocols for both S-CMR and CTP usually imply the administration of vasodilator agents to induce hyperemia and contrast agents to depict perfusion defects. However, both methods have their own limitations, meaning that optimizing their performance still requires a patient-tailored approach. This review focuses on the characteristics, drawbacks, and future perspectives of these two techniques.

Ensemble Deep Learning Derived from Transfer Learning for Classification of COVID-19 Patients on Hybrid Deep-Learning-Based Lung Segmentation: A Data Augmentation and Balancing Framework.

BACKGROUND AND MOTIVATION: Lung computed tomography (CT) techniques are high-resolution and are well adopted in the intensive care unit (ICU) for COVID-19 disease control classification. Most artificial intelligence (AI) systems do not undergo generalization and are typically overfitted. Such trained AI systems are not practical for clinical settings and therefore do not give accurate results when executed on unseen data sets. We hypothesize that ensemble deep learning (EDL) is superior to deep transfer learning (TL) in both non-augmented and augmented frameworks. METHODOLOGY: The system consists of a cascade of quality control, ResNet-UNet-based hybrid deep learning for lung segmentation, and seven models using TL-based classification followed by five types of EDL's. To prove our hypothesis, five different kinds of data combinations (DC) were designed using a combination of two multicenter cohorts-Croatia (80 COVID) and Italy (72 COVID and 30 controls)-leading to 12,000 CT slices. As part of generalization, the system was tested on unseen data and statistically tested for reliability/stability. RESULTS: Using the K5 (80:20) cross-validation protocol on the balanced and augmented dataset, the five DC datasets improved TL mean accuracy by 3.32%, 6.56%, 12.96%, 47.1%, and 2.78%, respectively. The five EDL systems showed improvements in accuracy of 2.12%, 5.78%, 6.72%, 32.05%, and 2.40%, thus validating our hypothesis. All statistical tests proved positive for reliability and stability. CONCLUSION: EDL showed superior performance to TL systems for both (a) unbalanced and unaugmented and (b) balanced and augmented datasets for both (i) seen and (ii) unseen paradigms, validating both our hypotheses.

Evaluation of operator eye exposure and eye protective devices in interventional radiology: Results on clinical staff and phantom.

PURPOSE: To assess occupational eye lens dose based on clinical monitoring of interventional radiologists and to assess personal protective eyewear (PPE) efficacy through measurements with anthropomorphic phantom. METHODS: Two positions of the operator with respect to X-ray beam were simulated with phantom. Dose reduction factor (DRF) of four PPE was assessed, as well as correlation between eye lens and whole-body doses. Brain dose was also assessed. Five radiologists were monitored for one-year clinical procedures. All subjects were equipped with whole-body dosimeter placed over lead apron at the chest level and eye lens dosimeter placed over the left side of the PPE. Kerma-Area Product (KAP) of procedures performed during the monitoring period was recorded. The correlation of eye lens dose with whole-body dose and KAP was assessed. RESULTS: DRF was 4.3/2.4 for wraparound glasses, 4.8/1.9 for fitover glasses, 9.1/6.8 for full-face visor in radial/femoral geometries. DRF of half-face visor depended on how it is worn (range 1.0-4.9). Statistically significant correlation between dose value over the PPE and chest dose was observed, while there was no correlation between eye lens dose and chest dose. The results on clinical staff showed statistically significant correlation between dose values over the PPE and KAP. CONCLUSIONS: All PPE showed significant DRF in all configurations, provided they were worn correctly. Single DRF value is not applicable to all clinical situations. KAP is a valuable tool for determining appropriate radiation protection measures.

T1-Weighted Contrast Enhancement, Apparent Diffusion Coefficient, and Cerebral-Blood-Volume Changes after Glioblastoma Resection: MRI within 48 Hours vs. beyond 48 Hours.

BACKGROUND: The aim of the study is to identify the advantages, if any, of post-operative MRIs performed at 48 h compared to MRIs performed after 48 h in glioblastoma surgery. MATERIALS AND METHODS: To assess the presence of a residual tumor, the T1-weighted Contrast Enhancement (CE), Apparent Diffusion Coefficient (ADC), and Cerebral Blood Volume (rCBV) in the proximity of the surgical cavity were considered. The rCBV ratio was calculated by comparing the rCBV with the contralateral normal white matter. After the blind image examinations by the two radiologists, the patients were divided into two groups according to time window after surgery: ≤48 h (group 1) and >48 h (group 2). RESULTS: A total of 145 patients were enrolled; at the 6-month follow-up MRI, disease recurrence was 89.9% (125/139), with a mean patient survival of 8.5 months (SD 7.8). The mean ADC and rCBV ratio values presented statistical differences between the two groups (p < 0.05). Of these 40 patients in whom an ADC value was not obtained, the rCBV values could not be calculated in 52.5% (21/40) due to artifacts (p < 0.05). CONCLUSION: The study showed differences in CE, rCBV, and ADC values between the groups of patients undergoing MRIs before and after 48 h. An MRI performed within 48 h may increase the ability of detecting GBM by the perfusion technique with the calculation of the rCBV ratio.

May an incidental finding on chest CT be a predictor of access in intensive care unit? Role of hepatic steatosis in patients affected by SARS-CoV-2.

Background: Due to the high prevalence of hepatic steatosis (HS), the aim of the study is to verify the frequency of HS incidentally detected in chest computed tomography (CT) imaging in our population affected by SARS-CoV-2 and to investigate its association with the severity of the infection and outcome in terms of hospitalization. Design and methods: We retrospectively analyzed 500 patients with flu syndrome and clinically suspected of having Sars-CoV-2 infection who underwent unenhanced chest CT and have positive RT-PCR tests for Sars-CoV-2 RNA. Two radiologists both with >5 years of thoracic imaging experience, evaluated the images in consensus, without knowing the RT-PCR results. Liver density was measured by a region of interest (ROI), using a liver attenuation value ≤40 Hounsfield units (HU). Results: On 480 patients, 23.1% (111/480) had an incidental findings of HS on chest CT. The steatosis group, included 83 (74.7%) males and 28 (25.3%) females. Patients with HS were more likely to be hospitalized in the intensive care unit (ICU). On univariate analysis, there is a correlation between probability to be intubate (access in the ICU) and HS: patients with HS are twice as likely to be intubated (OR 2.04, CI 95% 1.11-3.73). Conclusion: Chest CT is an important diagnostic tool for COVID-19 and can provide information about the prognosis of the disease. HS can easily be detected on chest CT taken for the diagnosis of the COVID-19 disease, is an important sign for a poor prognosis and possible predictor of admission in ICU.

Advanced imaging and Crohn's disease: An overview of clinical application and the added value of artificial intelligence.

PURPOSE: The purpose of this narrative review is to describe the clinical applications of advanced computed tomography (CT) and magnetic resonance (MRI) techniques in patients affected by Crohn's disease (CD), giving insights about the added value of artificial intelligence (AI) in this field. METHODS: We performed a literature search comparing standardized and advanced imaging techniques for CD diagnosis. Cross-sectional imaging is essential for the identification of lesions, the assessment of active or relapsing disease and the evaluation of complications. RESULTS: The studies reviewed show that new advanced imaging techniques and new MRI sequences could be integrated into standard protocols, to achieve a reliable quantification of CD activity, improve the lesions' characterization and the evaluation of therapy response. These promising tools are: dual-energy CT (DECT) post-processing techniques, diffusion-weighted MRI (DWI-MRI), dynamic contrast-enhanced MRI (DCE-MRI), Magnetization Transfer MRI (MT-MRI) and CINE-MRI. Furthermore, AI solutions show a potential when applied to radiological techniques in these patients. Machine learning (ML) algorithms and radiomic features prove to be useful in improving the diagnostic accuracy of clinicians and in attempting a personalized medicine approach, stratifying patients by predicting their prognosis. CONCLUSIONS: Advanced imaging is crucial in the diagnosis, lesions' characterisation and in the estimation of the abdominal involvement in CD. New AI developments are promising tools that could support doctors in the management of CD affected patients.

Non-timely clinically applicable ADC ratio in prostate mpMRI: a comparison with fusion biopsy results.

PURPOSE: The purpose of the study was to assess the diagnostic accuracy of ADC ratio and to evaluate its efficacy in reducing the number of false positives in prostatic mpMRI. MATERIALS AND METHODS: All patients who underwent an mpMRI and a targeted fusion biopsy in our institution from 2016 to 2021 were retrospectively selected. Two experienced readers (R1 and R2) independently evaluated the images, blindly to biopsy results. The radiologists assessed the ADC ratios by tracing a circular 10 mm2 ROI on the biopsied lesion and on the apparently benign contralateral parenchyma. Prostate cancers were divided into non-clinically significant (nsPC, Gleason score = 6) and clinically significant (sPC, Gleason score ≥ 7). ROC analyses were performed. RESULTS: 167 patients and188 lesions were included. Concordance was 0.62 according to Cohen's K. ADC ratio showed an AUC for PCAs of 0.78 in R1 and 0.8 in R2. The AUC for sPC was 0.85 in R1 and 0.84 in R2. The 100% sensitivity cut-off for sPCs was 0.65 (specificity 25.6%) in R1 and 0.66 (specificity 27.4%) in R2. Forty-three benign or not clinically significant lesions were above the 0.65 threshold in R1; 46 were above the 0.66 cut-off in R2. This would have allowed to avoid an equal number of unnecessary biopsies at the cost of 2 nsPCs in R1 and one nsPC in R2. CONCLUSION: In our sample, the ADC ratio was a useful and accurate tool that could potentially reduce the number of false positives in mpMRI.

Multicenter Study on COVID-19 Lung Computed Tomography Segmentation with varying Glass Ground Opacities using Unseen Deep Learning Artificial Intelligence Paradigms: COVLIAS 1.0 Validation.

Variations in COVID-19 lesions such as glass ground opacities (GGO), consolidations, and crazy paving can compromise the ability of solo-deep learning (SDL) or hybrid-deep learning (HDL) artificial intelligence (AI) models in predicting automated COVID-19 lung segmentation in Computed Tomography (CT) from unseen data leading to poor clinical manifestations. As the first study of its kind, "COVLIAS 1.0-Unseen" proves two hypotheses, (i) contrast adjustment is vital for AI, and (ii) HDL is superior to SDL. In a multicenter study, 10,000 CT slices were collected from 72 Italian (ITA) patients with low-GGO, and 80 Croatian (CRO) patients with high-GGO. Hounsfield Units (HU) were automatically adjusted to train the AI models and predict from test data, leading to four combinations-two Unseen sets: (i) train-CRO:test-ITA, (ii) train-ITA:test-CRO, and two Seen sets: (iii) train-CRO:test-CRO, (iv) train-ITA:test-ITA. COVILAS used three SDL models: PSPNet, SegNet, UNet and six HDL models: VGG-PSPNet, VGG-SegNet, VGG-UNet, ResNet-PSPNet, ResNet-SegNet, and ResNet-UNet. Two trained, blinded senior radiologists conducted ground truth annotations. Five types of performance metrics were used to validate COVLIAS 1.0-Unseen which was further benchmarked against MedSeg, an open-source web-based system. After HU adjustment for DS and JI, HDL (Unseen AI) > SDL (Unseen AI) by 4% and 5%, respectively. For CC, HDL (Unseen AI) > SDL (Unseen AI) by 6%. The COVLIAS-MedSeg difference was < 5%, meeting regulatory guidelines.Unseen AI was successfully demonstrated using automated HU adjustment. HDL was found to be superior to SDL.

Misophonia: Analysis of the neuroanatomic patterns at the basis of psychiatric symptoms and changes of the orthosympathetic/ parasympathetic balance.

Background/Aim: Misophonia is a disorder characterized by reduced tolerance to specific sounds or stimuli known as "triggers," which tend to evoke negative emotional, physiological, and behavioral responses. In this study, we aimed to better characterize participants with misophonia through the evaluation of the response of the autonomic nervous system to "trigger sounds," a psychometric assessment, and the analysis of the neurological pathways. Materials and methods: Participants included 11 adults presenting with misophonic disturbance and 44 sex-matched healthy controls (HCs). Following recently proposed diagnostic criteria, the participants listened to six "trigger sounds" and a "general annoyance" sound (baby crying) during a series of physiological tests. The effects were examined through functional magnetic resonance imaging (fMRI), the analysis of heart rate variability (HRV), and of galvanic skin conductance (GSC). The fMRI was performed on a 3T Scanner. The HRV was obtained through the analysis of electrocardiogram, whereas the GSC was examined through the positioning of silver-chloride electrodes on fingers. Furthermore, the psychometric assessment included questionnaires focused on misophonia, psychopathology, resilience, anger, and motivation. Results: Participants with misophonia showed patterns of increased sympathetic activation in response to trigger sounds and a general annoyance sound, the low frequency (LF) component of HRV, the sympathetic index, and the number of significant GSC over the threshold, where the amplitude/phasic response of GSC was higher. The fMRI analysis provided evidence for the activation of the temporal cortex, the limbic area, the ventromedial prefrontal/premotor/cingulate cortex, and the cerebellum in participants with misophonia. In addition, the psychometric assessment seemed to differentiate misophonia as a construct independent from general psychopathology. Conclusion: These results suggest the activation of a specific auditory-insula-limbic pathway at the basis of the sympathetic activation observed in participants with misophonia in response to "trigger and general annoyance sounds." Further studies should disentangle the complex issue of whether misophonia represents a new clinical disorder or a non-pathological condition. These results could help to build diagnostic tests to recognize and better classify this disorder. The relevance of this question goes beyond purely theoretical issues, as in the first case, participants with misophonia should receive a diagnosis and a targeted treatment, while in the second case, they should not.