-
BMC Medical Informatics and Decision... Aug 2023Differentiating between Crohn's disease (CD) and intestinal tuberculosis (ITB) with endoscopy is challenging. We aim to perform more accurate endoscopic diagnosis...
BACKGROUND
Differentiating between Crohn's disease (CD) and intestinal tuberculosis (ITB) with endoscopy is challenging. We aim to perform more accurate endoscopic diagnosis between CD and ITB by building a trustworthy AI differential diagnosis application.
METHODS
A total of 1271 electronic health record (EHR) patients who had undergone colonoscopies at Peking Union Medical College Hospital (PUMCH) and were clinically diagnosed with CD (n = 875) or ITB (n = 396) were used in this study. We build a workflow to make diagnoses with EHRs and mine differential diagnosis features; this involves finetuning the pretrained language models, distilling them into a light and efficient TextCNN model, interpreting the neural network and selecting differential attribution features, and then adopting manual feature checking and carrying out debias training.
RESULTS
The accuracy of debiased TextCNN on differential diagnosis between CD and ITB is 0.83 (CR F1: 0.87, ITB F1: 0.77), which is the best among the baselines. On the noisy validation set, its accuracy was 0.70 (CR F1: 0.87, ITB: 0.69), which was significantly higher than that of models without debias. We also find that the debiased model more easily mines the diagnostically significant features. The debiased TextCNN unearthed 39 diagnostic features in the form of phrases, 17 of which were key diagnostic features recognized by the guidelines.
CONCLUSION
We build a trustworthy AI differential diagnosis application for differentiating between CD and ITB focusing on accuracy, interpretability and robustness. The classifiers perform well, and the features which had statistical significance were in agreement with clinical guidelines.
Topics: Humans; Crohn Disease; Diagnosis, Differential; Tuberculosis, Gastrointestinal; Colonoscopy
PubMed: 37582768
DOI: 10.1186/s12911-023-02257-6 -
Computer Methods and Programs in... Oct 2022Nowadays, COVID-19 is spreading rapidly worldwide, and seriously threatening lives . From the perspective of security and economy, the effective control of COVID-19 has...
OBJECTIVE
Nowadays, COVID-19 is spreading rapidly worldwide, and seriously threatening lives . From the perspective of security and economy, the effective control of COVID-19 has a profound impact on the entire society. An effective strategy is to diagnose earlier to prevent the spread of the disease and prompt treatment of severe cases to improve the chance of survival.
METHODS
The method of this paper is as follows: Firstly, the collected data set is processed by chest film image processing, and the bone removal process is carried out in the rib subtraction module. Then, the set preprocessing method performed histogram equalization, sharpening, and other preprocessing operations on the chest film. Finally, shallow and high-level feature mapping through the backbone network extracts the processed chest radiographs. We implement the self-attention mechanism in Inception-Resnet, perform the standard classification, and identify chest radiograph diseases through the classifier to realize the auxiliary COVID-19 diagnosis process at the medical level, all in an effort to further enhance the classification performance of the convolutional neural network. Numerous computer simulations demonstrate that the Inception-Resnet convolutional neural network performs CT image categorization and enhancement with greater efficiency and flexibility than conventional segmentation techniques.
RESULTS
The experimental COVID-19 CT dataset obtained in this paper is the new data for CT scans and medical imaging of normal, early COVID-19 patients and severe COVID-19 patients from Jinyintan hospital. The experiment plots the relationship between model accuracy, model loss and epoch, using ACC, TPR, SPE, F1 score and G-mean to measure the image maps of patients with and without the disease. Statistical measurement values are obtained by Inception-Resnet are 88.23%, 83.45%, 89.72%, 95.53% and 88.74%. The experimental results show that Inception-Resnet plays a more effective role than other image classification methods in evaluation indicators, and the method has higher robustness, accuracy and intuitiveness.
CONCLUSION
With CT images in the clinical diagnosis of COVID-19 images being widely used and the number of applied samples continuously increasing, the method in this paper is expected to become an additional diagnostic tool that can effectively improve the diagnostic accuracy of clinical COVID-19 images.
Topics: COVID-19; COVID-19 Testing; Humans; Image Processing, Computer-Assisted; Lung; Neural Networks, Computer
PubMed: 35964421
DOI: 10.1016/j.cmpb.2022.107053 -
Journal of Translational Medicine Jan 2021Hysteroscopy is a commonly used technique for diagnosing endometrial lesions. It is essential to develop an objective model to aid clinicians in lesion diagnosis, as...
BACKGROUND
Hysteroscopy is a commonly used technique for diagnosing endometrial lesions. It is essential to develop an objective model to aid clinicians in lesion diagnosis, as each type of lesion has a distinct treatment, and judgments of hysteroscopists are relatively subjective. This study constructs a convolutional neural network model that can automatically classify endometrial lesions using hysteroscopic images as input.
METHODS
All histopathologically confirmed endometrial lesion images were obtained from the Shengjing Hospital of China Medical University, including endometrial hyperplasia without atypia, atypical hyperplasia, endometrial cancer, endometrial polyps, and submucous myomas. The study included 1851 images from 454 patients. After the images were preprocessed (histogram equalization, addition of noise, rotations, and flips), a training set of 6478 images was input into a tuned VGGNet-16 model; 250 images were used as the test set to evaluate the model's performance. Thereafter, we compared the model's results with the diagnosis of gynecologists.
RESULTS
The overall accuracy of the VGGNet-16 model in classifying endometrial lesions is 80.8%. Its sensitivity to endometrial hyperplasia without atypia, atypical hyperplasia, endometrial cancer, endometrial polyp, and submucous myoma is 84.0%, 68.0%, 78.0%, 94.0%, and 80.0%, respectively; for these diagnoses, the model's specificity is 92.5%, 95.5%, 96.5%, 95.0%, and 96.5%, respectively. When classifying lesions as benign or as premalignant/malignant, the VGGNet-16 model's accuracy, sensitivity, and specificity are 90.8%, 83.0%, and 96.0%, respectively. The diagnostic performance of the VGGNet-16 model is slightly better than that of the three gynecologists in both classification tasks. With the aid of the model, the overall accuracy of the diagnosis of endometrial lesions by gynecologists can be improved.
CONCLUSIONS
The VGGNet-16 model performs well in classifying endometrial lesions from hysteroscopic images and can provide objective diagnostic evidence for hysteroscopists.
Topics: China; Deep Learning; Endometrial Hyperplasia; Endometrial Neoplasms; Female; Humans; Hysteroscopy; Pregnancy; Sensitivity and Specificity; Uterine Diseases
PubMed: 33407588
DOI: 10.1186/s12967-020-02660-x -
MedEdPORTAL : the Journal of Teaching... 2023Cervical intraepithelial neoplasia 3 is associated with a high degree of progression to cervical cancer. Its risk is markedly reduced after excisional treatment. Hence,...
INTRODUCTION
Cervical intraepithelial neoplasia 3 is associated with a high degree of progression to cervical cancer. Its risk is markedly reduced after excisional treatment. Hence, it is critical that providers accurately diagnose and treat this condition. We present a simulation-based module focused on resident mastery of performance of colposcopy and loop electrosurgical excision procedure (LEEP).
METHODS
Learners were obstetrics and gynecology residents. Guidelines on performance of colposcopy and LEEP were presented prior to module participation. We used pelvic task trainers, kielbasa sausages, and routine equipment for performance of colposcopy and LEEP. Colposcopy and LEEP sessions each lasted 30 minutes. Learners completed questionnaires before and after regarding comfort level on aspects of colposcopy and LEEP performance and level of agreement with statements on performing procedures independently. Comfort levels and degrees of agreement were based on 5-point Likert scales (1 = 3 = 5 = respectively).
RESULTS
Modules were held in November 2021 and May 2022. Thirty-four residents participated. Mean comfort scores significantly increased from 3.1 to 4.3 ( < .001) before and after the module for all steps. There was an increase in level of agreement with statements on being able to independently perform colposcopy (2.2 to 3.5, < .01) and LEEP (2.9 to 3.6, = .06).
DISCUSSION
Simulation-based modules on performance of colposcopy and LEEP significantly increased resident learner comfort in the performance of these procedures. Comfort in performing these procedures is important in providing comprehensive gynecologic care.
Topics: Pregnancy; Female; Humans; Colposcopy; Electrosurgery; Computer Simulation; Obstetrics; Pelvis
PubMed: 37691878
DOI: 10.15766/mep_2374-8265.11344 -
Scientific Reports Mar 2021The reverse transcription-polymerase chain reaction (RT-PCR) assay is the accepted standard for coronavirus disease 2019 (COVID-19) diagnosis. As any test, RT-PCR...
The reverse transcription-polymerase chain reaction (RT-PCR) assay is the accepted standard for coronavirus disease 2019 (COVID-19) diagnosis. As any test, RT-PCR provides false negative results that can be rectified by clinicians by confronting clinical, biological and imaging data. The combination of RT-PCR and chest-CT could improve diagnosis performance, but this would requires considerable resources for its rapid use in all patients with suspected COVID-19. The potential contribution of machine learning in this situation has not been fully evaluated. The objective of this study was to develop and evaluate machine learning models using routine clinical and laboratory data to improve the performance of RT-PCR and chest-CT for COVID-19 diagnosis among post-emergency hospitalized patients. All adults admitted to the ED for suspected COVID-19, and then hospitalized at Rennes academic hospital, France, between March 20, 2020 and May 5, 2020 were included in the study. Three model types were created: logistic regression, random forest, and neural network. Each model was trained to diagnose COVID-19 using different sets of variables. Area under the receiving operator characteristics curve (AUC) was the primary outcome to evaluate model's performances. 536 patients were included in the study: 106 in the COVID group, 430 in the NOT-COVID group. The AUC values of chest-CT and RT-PCR increased from 0.778 to 0.892 and from 0.852 to 0.930, respectively, with the contribution of machine learning. After generalization, machine learning models will allow increasing chest-CT and RT-PCR performances for COVID-19 diagnosis.
Topics: Area Under Curve; COVID-19; COVID-19 Testing; Diagnosis, Computer-Assisted; Humans; Machine Learning; Proof of Concept Study; Reverse Transcriptase Polymerase Chain Reaction; Tomography, X-Ray Computed
PubMed: 33785852
DOI: 10.1038/s41598-021-86735-9 -
International Journal of Clinical... Nov 2022Sinusoidal obstruction syndrome (SOS) refers to liver injury caused by hematopoietic stem cell transplantation (HSCT) and anticancer drugs including oxaliplatin.... (Observational Study)
Observational Study
BACKGROUND
Sinusoidal obstruction syndrome (SOS) refers to liver injury caused by hematopoietic stem cell transplantation (HSCT) and anticancer drugs including oxaliplatin. Increased splenic volume (SV) on computed tomography (CT) indicates oxaliplatin-induced SOS. Similarly, ultrasonography and liver stiffness measurement (LSM) by shear-wave elastography (SWE) can help diagnose SOS after HSCT; however, their usefulness for diagnosing oxaliplatin-induced SOS remains unclear. We investigated the usefulness of the Hokkaido ultrasonography-based scoring system with 10 ultrasonographic parameters (HokUS-10) and SWE in diagnosing oxaliplatin-induced SOS early.
METHODS
In this prospective observational study, ultrasonography and SWE were performed before and at 2, 4, and 6 months after oxaliplatin-based chemotherapy. HokUS-10 was used for assessment. CT volumetry of the SV was performed in clinical practice, and an SV increase ≥ 30% was considered the diagnostic indicator of oxaliplatin-induced SOS. We assessed whether HokUS-10 and SWE can lead to an early detection of oxaliplatin-induced SOS before an increased SV on CT.
RESULTS
Of the 30 enrolled patients with gastrointestinal cancers, 12 (40.0%) with an SV increase ≥ 30% on CT were diagnosed with SOS. The HokUS-10 score was not correlated with an SV increase ≥ 30% (r = 0.18). The change in rate of three HokUS-10 parameters were correlated with an SV increase ≥ 30% (r = 0.32-0.41). The change in rate of LSM by SWE was correlated with an SV increase ≥ 30% (r = 0.40).
CONCLUSIONS
The usefulness of HokUS-10 score was not demonstrated; however, some HokUS-10 parameters and SWE could be useful for the early diagnosis of oxaliplatin-induced SOS.
Topics: Humans; Hepatic Veno-Occlusive Disease; Oxaliplatin; Elasticity Imaging Techniques; Ultrasonography; Antineoplastic Agents
PubMed: 36042137
DOI: 10.1007/s10147-022-02235-4 -
Frontiers in Pediatrics 2021Leukemia is the most common malignancy affecting children. The morphologic analysis of bone marrow smears is an important initial step for diagnosis. Recent publications...
Leukemia is the most common malignancy affecting children. The morphologic analysis of bone marrow smears is an important initial step for diagnosis. Recent publications demonstrated that artificial intelligence is able to classify blood cells but a long way from clinical use. A total of 1,732 bone marrow images were used for the training of a convolutional neural network (CNN). New techniques of deep learning were integrated and an end-to-end leukemia diagnosis system was developed by using raw images without pre-processing. The system creatively imitated the workflow of a hematologist by detecting and excluding uncountable and crushed cells, then classifying and counting the remain cells to make a diagnosis. The performance of the CNN in classifying WBCs achieved an accuracy of 82.93%, precision of 86.07% and F1 score of 82.02%. And the performance in diagnosing acute lymphoid leukemia achieved an accuracy of 89%, sensitivity of 86% and specificity of 95%. The system also performs well at detecting the bone marrow metastasis of lymphoma and neuroblastoma, achieving an average accuracy of 82.93%. This is the first study which included a wider variety of cell types in leukemia diagnosis, and achieved a relatively high performance in real clinical scenarios.
PubMed: 34249819
DOI: 10.3389/fped.2021.693676 -
Cancer Cytopathology Feb 2022Despite widespread clinical use, lymph node fine-needle aspiration cytology (LN-FNAC) lacks universal acceptance for definitively diagnosing lymphomas. This is likely...
BACKGROUND
Despite widespread clinical use, lymph node fine-needle aspiration cytology (LN-FNAC) lacks universal acceptance for definitively diagnosing lymphomas. This is likely due to reports of lower diagnostic performance, inconsistent terminology use in cytopathology diagnostic reports, and only limited data on the clinical implications of LN-FNAC diagnoses. Recently, a uniform LN-FNAC cytopathological diagnostic reporting system was proposed (the Sydney System). This study evaluated LN-FNAC diagnostic performance and risks of malignancy associated with the proposed diagnostic categories.
METHODS
LN-FNAC specimens obtained in 2018-2019, with and without concurrent core biopsy, to evaluate for suspected lymphoma were analyzed (n = 349). LN-FNAC diagnoses were compared with final diagnoses obtained via subsequent tissue biopsy and/or clinical assessment.
RESULTS
The mean patient age was 57.6 years, and 41% were female. LN-FNAC was the initial diagnostic test in 223 (63.9%), and it was used to evaluate for recurrence in 126 (36.1%). LN-FNAC diagnosed 202 hematological malignancies (57.9%), 23 nonhematological malignancies (6.6%), and 124 reactive processes (35.5%). Subsequent tissue biopsy was performed in 42 (12%). The risks of malignancy per diagnostic category were as follows: inadequate, 58.3%; benign, 6.4%; atypical, 69.2%; suspicious, 96.7%; and malignant, 99.3%. LN-FNAC demonstrated up to 96.3% sensitivity, 91.91% specificity, and 87.35% accuracy. Optimal specimen quality and the use of intradepartmental consultation reduced diagnostic error rates in FNA cases without concurrent core biopsy (P = .029 and P = .0002 respectively).
CONCLUSIONS
LN-FNAC is accurate and reliable for the diagnosis of lymphoma. Inadequate LN-FNAC samples should be resampled due to a significant associated risk of lymphoma. The diagnostic performance of LN-FNAC may be improved with good specimen quality and reviews by multiple pathologists. Understanding the risks of malignancy associated with LN-FNAC diagnostic categories will help to guide optimal patient management.
Topics: Biopsy, Fine-Needle; Biopsy, Large-Core Needle; Cytological Techniques; Female; Humans; Lymph Nodes; Male; Middle Aged; Neoplasms; Sensitivity and Specificity
PubMed: 34661975
DOI: 10.1002/cncy.22523 -
Frontiers in Oncology 2023Pancreatic cystic neoplasms are increasingly diagnosed with the development of medical imaging technology and people's self-care awareness. However, two of their...
BACKGROUND
Pancreatic cystic neoplasms are increasingly diagnosed with the development of medical imaging technology and people's self-care awareness. However, two of their sub-types, serous cystic neoplasms (SCN) and mucinous cystic neoplasms (MCN), are often misclassified from each other. Because SCN is primarily benign and MCN has a high rate of malignant transformation. Distinguishing SCN and MCN is challenging and essential.
PURPOSE
MRIs have many different modalities, complete with SCN and MCN diagnosis information. With the help of an artificial intelligence-based algorithm, we aimed to propose a multi-modal hybrid deep learning network that can efficiently diagnose SCN and MCN using multi-modality MRIs.
METHODS
A cross-modal feature fusion structure was innovatively designed, combining features of seven modalities to realize the classification of SCN and MCN. 69 Patients with multi-modalities of MRIs were included, and experiments showed performances of every modality.
RESULTS
The proposed method with the optimized settings outperformed all other techniques and human radiologists with high accuracy of 75.07% and an AUC of 82.77%. Besides, the proposed disentanglement method outperformed other fusion methods, and delayed contrast-enhanced T1-weighted MRIs proved most valuable in diagnosing SCN and MCN.
CONCLUSIONS
Through the use of a contemporary artificial intelligence algorithm, physicians can attain high performance in the complex challenge of diagnosing SCN and MCN, surpassing human radiologists to a significant degree.
PubMed: 37795452
DOI: 10.3389/fonc.2023.1181270 -
Ultrasound in Medicine & Biology May 2022We investigated the diagnostic performance of qualitative and quantitative ultrasound criteria for anterosuperior acetabular labral tears (ALTs). In all, 118 people with...
We investigated the diagnostic performance of qualitative and quantitative ultrasound criteria for anterosuperior acetabular labral tears (ALTs). In all, 118 people with ALTs (120 hips; case group) and 31 asymptomatic volunteers (42 hips; control group) at Peking University Third Hospital between August 2018 and November 2019 were consecutively included. The labral cleft, labral heterogeneous echogenicity, labral plump morphology, paralabral cyst and labral focal hyperechoic area were used as the qualitative criteria. The anterosuperior labral cross-section area (CSA) was measured as the quantitative criterion. The diagnostic utility of the quantitative and qualitative criteria were explored with magnetic resonance imaging as the diagnostic gold standard. Labral heterogeneous echogenicity was the most sensitive criterion for diagnosing ALTs (up to 80.00%), and the specificity of diagnosing ALTs with paralabral cysts, labral focal hyperechoic area and subcortical cysts of the femoral head and neck was as high as 90.48%-100%. The labral CSA in the case group was 0.27 cm (0.21-0.39 cm), which was significantly larger compared with the control group (0.18 cm [0.14-0.23 cm]; p < 0.001). The area under the receiver operating characteristic curve was 0.802 for diagnosing ALTs according to the labral CSA. The sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of the combined qualitative criteria for diagnosing ALTs were 90.00%, 71.43%, 90.00%, 71.43% and 85.19%, respectively. Labral heterogeneous echogenicity is a sensitive criterion for diagnosing ALTs, and paralabral cysts, labral focal hyperechoic, area and subcortical cysts of the femoral head and neck are specific criteria. The CSA of the anterosuperior acetabular labrum measured by ultrasound can be used as a quantitative criterion to diagnose ALTs. The combination of labral qualitative criteria provides higher sensitivity and accuracy in diagnosing ALTs.
Topics: Acetabulum; Arthroscopy; Cartilage, Articular; Humans; Magnetic Resonance Imaging; Retrospective Studies; Sensitivity and Specificity; Ultrasonography
PubMed: 35256224
DOI: 10.1016/j.ultrasmedbio.2022.01.016