New trend in artificial intelligence-based assistive technology for thoracic imaging.

Although there is no solid agreement for artificial intelligence (AI), it refers to a computer system with intelligence similar to that of humans. Deep learning appeared in 2006, and more than 10 years have passed since the third AI boom was triggered by improvements in computing power, algorithm development, and the use of big data. In recent years, the application and development of AI technology in the medical field have intensified internationally. There is no doubt that AI will be used in clinical practice to assist in diagnostic imaging in the future. In qualitative diagnosis, it is desirable to develop an explainable AI that at least represents the basis of the diagnostic process. However, it must be kept in mind that AI is a physician-assistant system, and the final decision should be made by the physician while understanding the limitations of AI. The aim of this article is to review the application of AI technology in diagnostic imaging from PubMed database while particularly focusing on diagnostic imaging in thorax such as lesion detection and qualitative diagnosis in order to help radiologists and clinicians to become more familiar with AI in thorax.

Clinical applications of artificial intelligence in liver imaging.

This review outlines the current status and challenges of the clinical applications of artificial intelligence in liver imaging using computed tomography or magnetic resonance imaging based on a topic analysis of PubMed search results using latent Dirichlet allocation. LDA revealed that "segmentation," "hepatocellular carcinoma and radiomics," "metastasis," "fibrosis," and "reconstruction" were current main topic keywords. Automatic liver segmentation technology using deep learning is beginning to assume new clinical significance as part of whole-body composition analysis. It has also been applied to the screening of large populations and the acquisition of training data for machine learning models and has resulted in the development of imaging biomarkers that have a significant impact on important clinical issues, such as the estimation of liver fibrosis, recurrence, and prognosis of malignant tumors. Deep learning reconstruction is expanding as a new technological clinical application of artificial intelligence and has shown results in reducing contrast and radiation doses. However, there is much missing evidence, such as external validation of machine learning models and the evaluation of the diagnostic performance of specific diseases using deep learning reconstruction, suggesting that the clinical application of these technologies is still in development.

Clinical Utility of Breast Ultrasound Images Synthesized by a Generative Adversarial Network.

BACKGROUND AND OBJECTIVES: This study compares the clinical properties of original breast ultrasound images and those synthesized by a generative adversarial network (GAN) to assess the clinical usefulness of GAN-synthesized images. MATERIALS AND METHODS: We retrospectively collected approximately 200 breast ultrasound images for each of five representative histological tissue types (cyst, fibroadenoma, scirrhous, solid, and tubule-forming invasive ductal carcinomas) as training images. A deep convolutional GAN (DCGAN) image-generation model synthesized images of the five histological types. Two diagnostic radiologists (reader 1 with 13 years of experience and reader 2 with 7 years of experience) were given a reading test consisting of 50 synthesized and 50 original images (≥1-month interval between sets) to assign the perceived histological tissue type. The percentages of correct diagnoses were calculated, and the reader agreement was assessed using the kappa coefficient. RESULTS: The synthetic and original images were indistinguishable. The correct diagnostic rates from the synthetic images for readers 1 and 2 were 86.0% and 78.0% and from the original images were 88.0% and 78.0%, respectively. The kappa values were 0.625 and 0.650 for the synthetic and original images, respectively. The diagnoses made from the DCGAN synthetic images and original images were similar. CONCLUSION: The DCGAN-synthesized images closely resemble the original ultrasound images in clinical characteristics, suggesting their potential utility in clinical education and training, particularly for enhancing diagnostic skills in breast ultrasound imaging.

Short- and long-term outcomes of immediate breast reconstruction surgery after neoadjuvant chemotherapy.

PURPOSE: Immediate breast reconstruction (IBR) is a standard option for breast cancer patients, although its utility in patients with advanced breast cancer requiring neoadjuvant chemotherapy (NAC) is debatable. We assessed the short-term complications and long-term prognosis of IBR after NAC. METHODS: We retrospectively analyzed 1135 patients with IBR and/or NAC between 2010 and 2018, 43 of whom underwent IBR after NAC. RESULTS: Twenty-five patients underwent reconstruction with a tissue expander (TE) followed by silicon breast implantation, 5 with a latissimus dorsi muscle transfer flap, and 13 with a deep inferior epigastric perforator flap. Complete surgical resection with a free margin confirmed by a pathological assessment was achieved in all patients. The evaluation of the short-term complications indicated no cases of total flap necrosis, two cases of partial flap necrosis, and one case of wound infection. Only one case required postponement of subsequent therapy due to partial flap necrosis. A long-term evaluation indicated no local recurrence, although distant metastasis was observed in 4 cases, 3 patients died, and TE removal after post-mastectomy radiotherapy (PMRT) was performed in 2 of 11 TE cases. CONCLUSION: IBR may be a viable option in patients with advanced breast cancer who achieve complete surgical resection after NAC.

Prognosis of Subcutaneous Mastectomy for Special Types of Breast Cancer.

Background and objectives: In the treatment of the special type of breast cancer (STBC), the choice of chemotherapeutic agents is often based on the characteristic features of the histological type. On the other hand, the surgical strategy is usually determined by the tumor size and presence of lymph node metastasis, and the indication for immediate reconstruction is rarely discussed based on the histological type. The prognoses of STBC and invasive ductal carcinoma of the breast (IDC) patients who underwent subcutaneous mastectomy (SCM) with immediate reconstruction at our institution were compared. Materials and Methods: A total of 254 patients with SCM with immediate reconstruction from 1998 to 2018 were included; their tumor diameter or induration was less than 25 mm, and it was not in close proximity to the skin. Preoperative chemotherapy and non-invasive cancer cases were excluded. Results: The number of patients was 166 for skin-sparing mastectomy (SSM) and 88 for nipple-sparing mastectomy (NSM). The reconstructive techniques were deep inferior epigastric artery perforator flap (DIEP) reconstruction in 43 cases, latissimus dorsi flap reconstruction (LDflap) in 63 cases, tissue expander (TE) in 117 cases, and transverse rectus abdominis myocutaneous flap/vertical rectus abdominis myocutaneous flap (TRAM/VRAM) reconstruction in 31 cases. The histological types of breast cancer were 211 IDC and 43 STBC; 17 were mucinous carcinoma (MUC), 17 were invasive lobular carcinoma (ILC), 6 were apocrine carcinoma, 1 was tubular carcinoma, and 2 were invasive micropapillary carcinoma. There was no difference in local recurrence or disease-free survival (LRFS, DFS) between IDC and STBC, and overall survival (OS) was significantly longer in STBC. OS was better in the STBC group because SCM with immediate reconstruction was performed for STBC, which is a histological type with a relatively good prognosis. Highly malignant histological types, such as squamous cell carcinoma or metaplastic carcinoma, were totally absent in this study. Conclusions: The indications for SCM with immediate reconstruction for relatively common STBCs such as MUC and ILC can be the same as for IDC.

Virtual Interpolation Images of Tumor Development and Growth on Breast Ultrasound Image Synthesis With Deep Convolutional Generative Adversarial Networks.

OBJECTIVES: We sought to generate realistic synthetic breast ultrasound images and express virtual interpolation images of tumors using a deep convolutional generative adversarial network (DCGAN). METHODS: After retrospective selection of breast ultrasound images of 528 benign masses, 529 malignant masses, and 583 normal breasts, 20 synthesized images of each were generated by the DCGAN. Fifteen virtual interpolation images of tumors were generated by changing the value of the input vector. A total of 60 synthesized images and 20 virtual interpolation images were evaluated by 2 readers, who scored them on a 5-point scale (1, very good; to 5, very poor) and then answered whether the synthesized image was benign, malignant, or normal. RESULTS: The mean score of overall quality for synthesized images was 3.05, and that of the reality of virtual interpolation images was 2.53. The readers classified the generated images with a correct answer rate of 92.5%. CONCLUSIONS: A DCGAN can generate high-quality synthetic breast ultrasound images of each pathologic tissue and has the potential to create realistic virtual interpolation images of tumor development.

Can Deep Learning-Based Volumetric Analysis Predict Oxygen Demand Increase in Patients with COVID-19 Pneumonia?

Background and Objectives: This study aimed to investigate whether predictive indicators for the deterioration of respiratory status can be derived from the deep learning data analysis of initial chest computed tomography (CT) scans of patients with coronavirus disease 2019 (COVID-19). Materials and Methods: Out of 117 CT scans of 75 patients with COVID-19 admitted to our hospital between April and June 2020, we retrospectively analyzed 79 CT scans that had a definite time of onset and were performed prior to any medication intervention. Patients were grouped according to the presence or absence of increased oxygen demand after CT scan. Quantitative volume data of lung opacity were measured automatically using a deep learning-based image analysis system. The sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) of the opacity volume data were calculated to evaluate the accuracy of the system in predicting the deterioration of respiratory status. Results: All 79 CT scans were included (median age, 62 years (interquartile range, 46-77 years); 56 (70.9%) were male. The volume of opacity was significantly higher for the increased oxygen demand group than for the nonincreased oxygen demand group (585.3 vs. 132.8 mL, p < 0.001). The sensitivity, specificity, and AUC were 76.5%, 68.2%, and 0.737, respectively, in the prediction of increased oxygen demand. Conclusion: Deep learning-based quantitative analysis of the affected lung volume in the initial CT scans of patients with COVID-19 can predict the deterioration of respiratory status to improve treatment and resource management.

Clinical Usefulness of Ultrasound-Guided Fine Needle Aspiration and Core Needle Biopsy for Patients with Axillary Lymphadenopathy.

Background and Objectives: It is necessary to properly diagnose and manage axillary lymphadenopathy caused by a variety of diseases. This study aimed to evaluate the utility of ultrasound (US)-guided sampling in patients with axillary lymphadenopathy. Materials and Methods: Patients with axillary lymphadenopathy (excluding patients with newly diagnosed breast cancer) who underwent US-guided fine needle aspiration (FNA) or core needle biopsy (CNB) at a single center between February 2016 and September 2020 were retrospectively examined. The association between US imaging findings and malignancy was investigated and the diagnostic performance of US-guided sampling was assessed. Results: Fifty-five patients (including eight males) were included in the study; of these, 34 patients (61.8%) were finally diagnosed with a malignant lymph node lesion. Twenty-two patients (40.0%) had undergone FNA and 33 (60.0%) had undergone CNB. Larger short and long axis diameters, thicker lymph node cortex, and the absence of fatty hilum on the US were significantly associated with malignancy (p < 0.05). The diagnostic performance of FNA, CNB, and FNA + CNB was excellent (sensitivity, specificity, and accuracy of 0.909, 0.900, and 0.917 for FNA, 0.958, 1.000, and 0.970 for CNB, and 0.941, 0.952, and 0.945 for FNA + CNB, respectively). Conclusions: US-guided FNA and CNB play an important role in the diagnosis and management of patients with axillary lymphadenopathy.

Virtual Navigator Real-Time Ultrasound Fusion Imaging with Positron Emission Tomography/Computed Tomography for Preoperative Breast Cancer.

We used virtual navigator real-time ultrasound (US) fusion imaging with 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) to identify a lesion that could not be detected on the US alone in a preoperative breast cancer patient. Of the patient's two lesions of breast cancer, the calcified lesion could not be identified by US alone. By fusing US with 18F-FDG PET/CT, which had been performed in advance, the location of the lesion could be estimated and marked, which benefited planning an appropriate surgery. The fusion of US and 18F-FDG PET/CT was a simple and noninvasive method for identifying the lesions detected by 18F-FDG PET/CT.

Classification of Breast Masses on Ultrasound Shear Wave Elastography using Convolutional Neural Networks.

We aimed to use deep learning with convolutional neural networks (CNNs) to discriminate images of benign and malignant breast masses on ultrasound shear wave elastography (SWE). We retrospectively gathered 158 images of benign masses and 146 images of malignant masses as training data for SWE. A deep learning model was constructed using several CNN architectures (Xception, InceptionV3, InceptionResNetV2, DenseNet121, DenseNet169, and NASNetMobile) with 50, 100, and 200 epochs. We analyzed SWE images of 38 benign masses and 35 malignant masses as test data. Two radiologists interpreted these test data through a consensus reading using a 5-point visual color assessment (SWEc) and the mean elasticity value (in kPa) (SWEe). Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated. The best CNN model (which was DenseNet169 with 100 epochs), SWEc, and SWEe had a sensitivity of 0.857, 0.829, and 0.914 and a specificity of 0.789, 0.737, and 0.763 respectively. The CNNs exhibited a mean AUC of 0.870 (range, 0.844-0.898), and SWEc and SWEe had an AUC of 0.821 and 0.855. The CNNs had an equal or better diagnostic performance compared with radiologist readings. DenseNet169 with 100 epochs, Xception with 50 epochs, and Xception with 100 epochs had a better diagnostic performance compared with SWEc (P = 0.018-0.037). Deep learning with CNNs exhibited equal or higher AUC compared with radiologists when discriminating benign from malignant breast masses on ultrasound SWE.

[A Case of Axillary Accessory Breast Cancer Treated via Latissimus Dorsi Musculocutaneous Flap Reconstruction].

A 46-year-old woman visited our hospital with the chief complaint of left axillary mass enlargement, which she had been aware of for 8 years. Palpation revealed that the mass was 15mm in size. Redness and gathering of the skin were also observed. Mammographic imaging of the left axilla revealed an irregular mass with skin infiltration. Breast ultrasonography revealed a low echo mass in the left axilla, which was continuous from the skin. Core needle biopsy was used to diagnose the tumor as an invasive ductal carcinoma. No other lesions were observed in the breast, and primary lesions were not found in any other organs. The patient was diagnosed with axillary accessory breast cancer and underwent local extensive resection and axillary lymph node dissection. Because the skin defect was widespread, we performed axillary reconstruction using the latissimus dorsi musculocutaneous flap to prevent upper limb contracture. At present, she can move her upper limbs and lymphedema has not been observed. In cases of axillary accessory breast cancer with skin infiltration, reconstruction using the latissimus dorsi musculocutaneous flap can be a useful procedure.

Crohn Disease: A 5-Point MR Enterocolonography Classification Using Enteroscopic Findings.

OBJECTIVE: The objectives of our study were to establish the efficacy of a 5-point MR enterocolonography classification for assessing Crohn disease (CD) activity, compare this classification with a validated MRI score (i.e., the MR index of activity [MaRIA]), and compare both with endoscopic findings, which were assessed using the Crohn disease endoscopic index of severity (CDEIS). MATERIALS AND METHODS: Seventy (derivation cohort) and 50 (validation cohort) patients with CD were retrospectively enrolled in this study. We developed a 5-point MR enterocolonography classification that consists of visual assessments alone. MR enterocolonography results were evaluated for each bowel segment (rectum; sigmoid, descending, transverse, and ascending colon; terminal and proximal ileum; and jejunum) by one observer in the derivation phase and independently by three observers in the validation phase using the 5-point MR enterocolonography classification lexicon and MaRIA. Areas under the ROC curves (AUCs) in discriminating endoscopic deep ulcers were compared between the MR enterocolonography classification and MaRIA. Interobserver reproducibility was assessed using weighted kappa coefficients. RESULTS: The AUCs of the MR enterocolonography classification were 89.0% in the derivation phase and 88.5%, 81.0%, and 77.3% for the three observers in the validation phase. The AUCs of the MR enterocolonography classification were statistically noninferior to those of MaRIA (p < 0.001). The cross-validation accuracy was 81.9% in the derivation phase and 81.5% in the validation phase. The MR enterocolonography classification showed good reproducibility. CONCLUSION: The 5-point MR enterocolonography classification was shown to be effective for evaluating CD activity in the large and small bowel.

[A Case of Breast Cancer Complicated with Castleman's Disease].

The patient was a 57-year-old woman with Castleman's disease. The follow-up CT scans obtained during the treatment of Castleman's disease, detected a 15mm nodule in the right breast AC area, and the patient was diagnosed with breast cancer. Lymphadenopathy was noted on both sides of the axilla; however, it was considered to be due to Catsleman's disease. Mastectomy and sentinel lymph node biopsy were performed with preoperative diagnosis of cT1cN0M0, cStage Ⅰ. Rapid diagnosis of the sentinel lymph node during the operation showed a metastatic tumor measuring 3mm and axillary dissection was performed. However, no metastasis was found in the dissected lymph node, which was, therefore, considered as an enlargement due to Castleman's disease.

Utility of Magnetic Resonance Enterography For Small Bowel Endoscopic Healing in Patients With Crohn's Disease.

OBJECTIVES: Small bowel (SB) endoscopic healing has not been well studied in patients with Crohn's disease (CD). This study aims to evaluate the utility of magnetic resonance (MR) enterography (MRE) for SB lesions in comparison with balloon-assisted enteroscopy (BAE) findings. METHODS: In total, 139 patients with CD in clinical-serological remission were prospectively followed after BAE and MRE procedures. We applied a modified version of the Simple Endoscopic Score for CD (SES-CD) for an endoscopic evaluation of the SB, called the Simple Endoscopic Active Score for CD (SES-CDa). We also used the MR index of activity (MaRIA) for MR evaluations. The primary end points were time to clinical relapse (CD activity index of >150 with an increase of >70 points) and serological relapse (abnormal elevation of C-reactive protein). RESULTS: Clinical and serological relapses occurred in 30 (21.6%) and 62 (44.6%) patients, respectively. SB endoscopic healing (SES-CDa<5) was observed in 76 (54.7%) patients. A multiple regression analysis showed that the lack of SB endoscopic healing was an independent risk factor for clinical relapses (hazard ratio (HR): 5.34; 95% confidence interval (CI): 2.06-13.81) and serological relapses (HR: 3.02; 95% CI: 1.65-5.51), respectively. MR ulcer healing (MaRIA score <11) demonstrated a high diagnostic accuracy (90.9%; 95% CI: 87.9-93.2%) for endoscopic healing. The kappa coefficient between BAE and MRE for longitudinal responsiveness was 0.754 (95% CI: 0.658-0.850) for clinical relapse and 0.783 (95% CI: 0.701-0.865) for serological relapse. CONCLUSIONS: SB inflammation was associated with a poor prognosis in patients with clinical-serological remission. MRE is a valid and reliable examination for SB inflammatory activity both for cross-sectional evaluations and prognostic prediction.

Clinical usefulness of 2-deoxy-2-[18F] fluoro-d-glucose-positron emission tomography/computed tomography for assessing early oral squamous cell carcinoma (cT1-2N0M0).

BACKGROUND: Positron emission tomography with 2-deoxy-2-[18F] fluoro-d-glucose integrated with computed tomography (FDG-PET/CT) is a useful method to evaluate patients with oral squamous cell carcinoma (OSCC). However, the prognostic significance of FDG-PET/CT for assessing early OSCC remains unclear. METHODS: Pretreatment FDG-PET/CT of 205 consecutive patients (125 men, 80 women, mean age 59.7 year old) with early OSCC (cT1-2N0M0) between June 2010 and December 2014 were retrospectively analyzed. FDG avidity in primary lesions was assessed by visual interpretation. Thereafter, maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV) and total lesion glycolysis (TLG) were measured in primary lesions. The relationship between each parameter and recurrence free survival (RFS) was assessed using the log-rank test. The performance of FDG-PET/CT for diagnosing metastatic lesions and synchronous cancer was also assessed. RESULTS: During the follow-up period (mean 32.9 months), 43 patients developed recurrences (21.0%). Patients with visually positive FDG uptake in primary lesions showed significantly shorter RFS than the others (63.0 months vs. 52.9 months, P = 0.005). In those patients, greater SUVmax, MTV, and TLG did not significantly predict shorter RFS. The sensitivity and specificity of FDG-PET/CT for cervical nodal metastases detection were 32.3% and 77.6%, respectively. FDG-PET/CT detected eight synchronous cancers (3.9%) and overlooked six synchronous cancers (2.9%). CONCLUSIONS: Although its utility for detecting cervical nodal metastases and synchronous cancers is limited, FDG-PET/CT is a potentially prognostic indicator in early OSCC.

[A Case of Ipsilateral Breast Tumor Recurrence with Breast Ulcer].

The patient was a 61-year-old woman with a history of treatment for breast cancer. She had undergone left breast-conserving surgery and axillary lymph node dissection followed by radiation therapy for left breast cancer when she was 43 years old and later presented to our hospital with a chief complaint of a left breast ulcer with a messy exudate. Core needle biopsy (CNB)from the ulcer showed no malignancy. PET-CT showed low FDG uptake at the left breast and high FDG uptake at a contralateral axillary lymph node. CNB of a contralateral axillary lymph node showed metastasis of breast cancer. She underwent left mastectomy and right axillary lymph node dissection for local control. Ipsilateral breast tumor recurrence in the left breast and contralateral axillary metastases were identified in the final pathological findings. The patient was treated with FEC 100 and tri-weekly docetaxel and has been recurrence-free for 2 years after surgery.

Clinical Response of Primary Malignant Pericardial Mesothelioma with Peritoneal Dissemination to Nivolumab.

Malignant pericardial mesothelioma (MPM) is extremely rare, and peritoneal dissemination has not yet been reported. There is no consensus regarding appropriate pharmacological treatment for MPM, including immune checkpoint inhibitors (ICIs). We herein report a 36-year-old man with MPM diagnosed by peritoneal dissemination and treated with an ICI. Cytology of the ascites revealed malignant peritonitis, and a re-evaluation of a pericardial biopsy performed at the previous hospital led to a diagnosis of MPM. The patient was treated with nivolumab and showed a clinical response despite several complications, such as renal dysfunction and performance status deterioration. This case report provides suggestive information for the diagnosis and ICI therapy of a rare type of mesothelioma.

A New Deep Learning Algorithm for Detecting Spinal Metastases on Computed Tomography Images.

STUDY DESIGN: Retrospective diagnostic study. OBJECTIVE: To automatically detect osteolytic bone metastasis lesions in the thoracolumbar region using conventional computed tomography (CT) scans, we developed a new deep learning (DL)-based computer-aided detection model. SUMMARY OF BACKGROUND DATA: Radiographic detection of bone metastasis is often difficult, even for orthopedic surgeons and diagnostic radiologists, with a consequent risk for pathologic fracture or spinal cord injury. If we can improve detection rates, we will be able to prevent the deterioration of patients' quality of life at the end stage of cancer. MATERIALS AND METHODS: This study included CT scans acquired at Tokyo Medical and Dental University (TMDU) Hospital between 2016 and 2022. A total of 263 positive CT scans that included at least one osteolytic bone metastasis lesion in the thoracolumbar spine and 172 negative CT scans without bone metastasis were collected for the datasets to train and validate the DL algorithm. As a test data set, 20 positive and 20 negative CT scans were separately collected from the training and validation datasets. To evaluate the performance of the established artificial intelligence (AI) model, sensitivity, precision, F1-score, and specificity were calculated. The clinical utility of our AI model was also evaluated through observer studies involving six orthopaedic surgeons and six radiologists. RESULTS: Our AI model showed a sensitivity, precision, and F1-score of 0.78, 0.68, and 0.72 (per slice) and 0.75, 0.36, and 0.48 (per lesion), respectively. The observer studies revealed that our AI model had comparable sensitivity to orthopaedic or radiology experts and improved the sensitivity and F1-score of residents. CONCLUSION: We developed a novel DL-based AI model for detecting osteolytic bone metastases in the thoracolumbar spine. Although further improvement in accuracy is needed, the current AI model may be applied to current clinical practice. LEVEL OF EVIDENCE: Level III.

Arterial Calcification Disappearance in Breast Imaging: A Key Indicator for Transition to Invasive Ductal Carcinoma.

A woman in her 70s, initially suspected of having fibroadenoma due to a well-defined mass in her breast, underwent regular mammography and ultrasound screenings. Over several years, no appreciable alterations in the mass were observed, maintaining the fibroadenoma diagnosis. However, in the fourth year, an ultrasound indicated slight enlargement and peripheral irregularities in the mass, even though the mammography images at that time showed no alterations. Interestingly, mammography images over time showed the gradual disappearance of previously observed arterial calcification around the mass. Pathological examination eventually identified the mass as invasive ductal carcinoma. Although the patient had breast tissue arterial calcification typical of atherosclerosis, none was present around the tumor-associated arteries. This case highlights the importance of monitoring arterial calcification changes in mammography, suggesting that they are crucial indicators in breast cancer diagnosis, beyond observing size and shape alterations.