Computer-aided analysis of radiological images for cancer diagnosis: performance analysis on benchmark datasets, challenges, and directions

Radiological image analysis using machine learning has been extensively applied to enhance biopsy diagnosis accuracy and assist radiologists with precise cures. With improvements in the medical industry and its technology, computer-aided diagnosis (CAD) systems have been essential in detecting early cancer signs in patients that could not be observed physically, exclusive of introducing errors. CAD is a detection system that combines artificially intelligent techniques with image processing applications thru computer vision. Several manual procedures are reported in state of the art for cancer diagnosis. Still, they are costly, time-consuming and diagnose cancer in late stages such as CT scans, radiography, and MRI scan. In this research, numerous state-of-the-art approaches on multi-organs detection using clinical practices are evaluated, such as cancer, neurological, psychiatric, cardiovascular and abdominal imaging. Additionally, numerous sound approaches are clustered together and their results are assessed and compared on benchmark datasets. Standard metrics such as accuracy, sensitivity, specificity and false-positive rate are employed to check the validity of the current models reported in the literature. Finally, existing issues are highlighted and possible directions for future work are also suggested.

Implications of introducing case based radiological images in anatomy on teaching, learning and assessment of medical students: a mixed-methods study

BACKGROUND

Introducing radiological anatomy in the preclinical curriculum can increase the understanding of Anatomy. Regardless of the integration when teaching anatomy, it is essential to maintain oversight as to what and how much is being taught. In addition, the knowledge requirements for preclinical students should be considered. The purpose of this kind of integration is that the student should be able to apply the knowledge which can help them better understand anatomy and not to make the course more challenging. This study aimed to understand whether adding radiological images would increase the difficulty level of the questions.

METHODS

We introduced radiological images, including X Rays, CT scans and MRIs, when teaching anatomy in the preclinical curriculum. A class of 99 students were tested using A-type MCQs (n = 84). All 84 questions were categorized on whether they were case-based with or without a radiological image. The item analysis of both groups of test questions was then compared based on their difficulty and discrimination index. A qualitative student perception regarding the inclusion of radiological images in anatomy was also measured using a questionnaire with a 5-point Likert scale.

RESULTS

The results showed that the performance level of the students was similar when comparing the test questions in both groups. The item analysis of the MCQs in the two groups revealed that by integrating radiological images when teaching anatomy, the various parameters in both groups of test questions were in the same range. More than 80% of the students felt that radiological images facilitate the achievement of learning outcomes and help to apply their knowledge in clinical contexts. The study's findings reported that the rate of satisfaction by including radiological images when teaching anatomy is high.

CONCLUSION

Recognition and interpretation of images are essential in an undergraduate medical program. Students found it helpful when radiological images were introduced to them when teaching anatomy. Since the students' performance in summative exams in both groups of questions was in the same range, the findings also point out that adding radiological images when teaching anatomy does not increase the difficulty of the subject.

Multi-modal fusion of deep transfer learning based COVID-19 diagnosis and classification using chest x-ray images

COVID-19 pandemic has a significant impact on the global health and daily lives of people living over the globe. Several initial tests are based on the detecting of the genetic material of the coronavirus, and they have a minimum detection rate with a time-consuming process. To overcome this issue, radiological images are recommended where chest X-rays (CXRs) are employed in the diagnostic process. This article introduces a new Multi-modal fusion of deep transfer learning (MMF-DTL) technique to classify COVID-19. The proposed MMF-DTL model involves three main processes, namely pre-processing, feature extraction, and classification. The MMF-DTL model uses three DL models namely VGG16, Inception v3, and ResNet 50 for feature extraction. Since a single modality would not be adequate to attain an effective detection rate, the integration of three approaches by the use of decision-based multimodal fusion increases the detection rate. So, a fusion of three DL models takes place to further improve the detection rate. Finally, a softmax classifier is employed for test images to a set of six different. A wide range of experimental result analyses is carried out on the Chest-X-Ray dataset. The proposed fusion model is found to be an effective tool for COVID-19 diagnosis using radiological images with the average sens _y of 92.96%, spec _y of 98.54%, prec _n of 93.60%, accu _y of 98.80%, F _score of 93.26% and kappa of 91.86%.

Solitary fibrous tumor of the pleura mimicking a soft tissue sarcoma of the chest wall

Introduction and importance

Solitary fibrous tumors of the pleura (SFTPs) present a diagnostic challenge. We herein report a successful case mimicking a soft tissue sarcoma of the chest wall by a meticulous evaluation of the conventional images.

Case presentation

A 51-year-old woman presented with a left thoracic mass. The mass exhibited an extrapleural sign, which suggested a chest wall origin. However, the mass was found to be located more caudally by additional computed tomography. This positional change suggested that the mass was pedunculated from the visceral pleura, and an SFTP was suspected. The mass was found to originate from the visceral pleura of the left lower lobe and a pathological diagnosis of an SFTP was confirmed.

Clinical discussion

Although a positional shift with a postural change or the respiratory phase is a well-known characteristic radiological finding, such an intentional imaging study is available only for suspicious cases of SFTPs.

Conclusions

SFTPs pose a diagnostic challenge because of their rarity and the lack of specific radiological findings. Even conventional radiological images can be diagnostic by performing a meticulous evaluation regardless of any specific diagnosis being initially assumed.

•

Solitary Fibrous Tumor of the Pleura (SFTP) poses a diagnostic challenge because of their rarity and the lack of specific radiological findings.

•

The positional shift with a postural change or the respiratory phase is one of the most characteristic radiological findings, but would be feasible only in the suspicious cases of SFTP.

•

Even conventional radiological images can be diagnostic by performing a meticulous evaluation regardless of any specific diagnosis being initially assumed.

Automatic detection of COVID-19 using pruned GLCM-Based texture features and LDCRF classification

Recently, automatic computer-aided detection (CAD) of COVID-19 using radiological images has received a great deal of attention from many researchers and medical practitioners, and consequently several CAD frameworks and methods have been presented in the literature to assist the radiologist physicians in performing diagnostic COVID-19 tests quickly, reliably and accurately. This paper presents an innovative framework for the automatic detection of COVID-19 from chest X-ray (CXR) images, in which a rich and effective representation of lung tissue patterns is generated from the gray level co-occurrence matrix (GLCM) based textural features. The input CXR image is first preprocessed by spatial filtering along with median filtering and contrast limited adaptive histogram equalization to improve the CXR image's poor quality and reduce image noise. Automatic thresholding by the optimized formula of Otsu's method is applied to find a proper threshold value to best segment lung regions of interest (ROIs) out from CXR images. Then, a concise set of GLCM-based texture features is extracted to accurately represent the segmented lung ROIs of each CXR image. Finally, the normalized features are fed into a trained discriminative latent-dynamic conditional random fields (LDCRFs) model for fine-grained classification to divide the cases into two categories: COVID-19 and non-COVID-19. The presented method has been experimentally tested and validated on a relatively large dataset of frontal CXR images, achieving an average accuracy, precision, recall, and F1-score of 95.88%, 96.17%, 94.45%, and 95.79%, respectively, which compare favorably with and occasionally exceed those previously reported in similar studies in the literature.

Data-efficient deep learning of radiological image data for outcome prediction after endovascular treatment of patients with acute ischemic stroke

Treatment selection is becoming increasingly more important in acute ischemic stroke patient care. Clinical variables and radiological image biomarkers (old age, pre-stroke mRS, NIHSS, occlusion location, ASPECTS, among others) have an important role in treatment selection and prognosis. Radiological biomarkers require expert annotation and are subject to inter-observer variability. Recently, Deep Learning has been introduced to reproduce these radiological image biomarkers. Instead of reproducing these biomarkers, in this work, we investigated Deep Learning techniques for building models to directly predict good reperfusion after endovascular treatment (EVT) and good functional outcome using CT angiography images. These models do not require image annotation and are fast to compute. We compare the Deep Learning models to Machine Learning models using traditional radiological image biomarkers. We explored Residual Neural Network (ResNet) architectures, adapted them with Structured Receptive Fields (RFNN) and auto-encoders (AE) for network weight initialization. We further included model visualization techniques to provide insight into the network's decision-making process. We applied the methods on the MR CLEAN Registry dataset with 1301 patients. The Deep Learning models outperformed the models using traditional radiological image biomarkers in three out of four cross-validation folds for functional outcome (average AUC of 0.71) and for all folds for reperfusion (average AUC of 0.65). Model visualization showed that the arteries were relevant features for functional outcome prediction. The best results were obtained for the ResNet models with RFNN. Auto-encoder initialization often improved the results. We concluded that, in our dataset, automated image analysis with Deep Learning methods outperforms radiological image biomarkers for stroke outcome prediction and has the potential to improve treatment selection.

Trichorhinophalangeal syndrome

Trichorhinophalangeal syndrome I (TPRSI) has an autosomal dominant inheritance; the proportion of «de novo» cases is unknown¹. It is characterized by unique facial features, bulbous nose, flat and elongated nasolabial furrow, thin hair and slow growth. Skeletal abnormalities that include short phalanges and metacarpals -brachydactyly-, cone-shaped epiphyses, hip dysplasia and short stature^1-3.

Spondylocostal Dysostosis (Jarcho-Levin Syndrome) in an Adult Patient with Consanguineous Parents, in Long-Term Follow-Up

A 24-year-old woman presented to neurosurgical consultation for chronic back pain. The patient was long term in wheelchair for vertebral deformity. She was the third child of first-degree consanguineous parents. The 2 older brothers had also vertebral malformations. The radiological images showed butterfly vertebra, vertebral fusion, hemivertebrae, scoliosis, and rib malformation. The patient was in follow-up for restrictive lung disease. Motor evoked potentials and lower limb electromyography were normal. We recommended conservative treatment for the back pain with antalgic and physical therapy. Diagnosis of spondylocostal dysostosis, or Jarcho-Levin syndrome, was made based on radiological features. Radiological mages are pathognomonic. Spondylocostal dysostosis is a rare hereditary disorder associated with multiple vertebral and rib anomalies. The entity is distinct from spondylothoracic dysostosis, which has a higher mortality due to respiratory complications. The patient was not compliant for genetic familiar counseling. At 12-year follow-up, the patient was in periodic respiratory and motor rehabilitation therapy.