Using the K-nearest neighbor algorithm for the classification of lymph node metastasis in gastric cancer. Comput Math Methods Med. 2012;2012:876545. [PMID: 23150740 PMCID: PMC3488413 DOI: 10.1155/2012/876545]

Comparative assessment of the capability of machine learning-based radiomic models for predicting omental metastasis in locally advanced gastric cancer

The study aims to investigate the predictive capability of machine learning algorithms for omental metastasis in locally advanced gastric cancer (LAGC) and to compare the performance metrics of various machine learning predictive models. A retrospective collection of 478 pathologically confirmed LAGC patients was undertaken, encompassing both clinical features and arterial phase computed tomography images. Radiomic features were extracted using 3D Slicer software. Clinical and radiomic features were further filtered through lasso regression. Selected clinical and radiomic features were used to construct omental metastasis predictive models using support vector machine (SVM), decision tree (DT), random forest (RF), K-nearest neighbors (KNN), and logistic regression (LR). The models' performance metrics included accuracy, area under the curve (AUC) of the receiver operating characteristic curve, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). In the training cohort, the RF predictive model surpassed LR, SVM, DT, and KNN in terms of accuracy, AUC, sensitivity, specificity, PPV, and NPV. Compared to the other four predictive models, the RF model significantly improved PPV. In the test cohort, all five machine learning predictive models exhibited lower PPVs. The DT model demonstrated the most significant variation in performance metrics relative to the other models, with a sensitivity of 0.231 and specificity of 0.990. The LR-based predictive model had the lowest PPV at 0.210, compared to the other four models. In the external validation cohort, the performance metrics of the predictive models were generally consistent with those in the test cohort. The LR-based model for predicting omental metastasis exhibited a lower PPV. Among the machine learning algorithms, the RF predictive model demonstrated higher accuracy and improved PPV relative to LR, SVM, KNN, and DT models.

Artificial intelligence in screening and diagnosis of surgical diseases: A narrative review

Artificial intelligence (AI) is playing an increasing role in several fields of medicine. It is also gaining popularity among surgeons as a valuable screening and diagnostic tool for many conditions such as benign and malignant colorectal, gastric, thyroid, parathyroid, and breast disorders. In the literature, there is no review that groups together the various application domains of AI when it comes to the screening and diagnosis of main surgical diseases. The aim of this review is to describe the use of AI in these settings. We performed a literature review by searching PubMed, Web of Science, Scopus, and Embase for all studies investigating the role of AI in the surgical setting, published between January 01, 2000, and June 30, 2023. Our focus was on randomized controlled trials (RCTs), meta-analysis, systematic reviews, and observational studies, dealing with large cohorts of patients. We then gathered further relevant studies from the reference list of the selected publications. Based on the studies reviewed, it emerges that AI could strongly enhance the screening efficiency, clinical ability, and diagnostic accuracy for several surgical conditions. Some of the future advantages of this technology include implementing, speeding up, and improving the automaticity with which AI recognizes, differentiates, and classifies the various conditions.

Pre- and Post-treatment Double-Sequential-Point Dynamic Radiomic Model in the Response Prediction of Gastric Cancer to Neoadjuvant Chemotherapy: 3-Year Survival Analysis

BACKGROUND

Prognosis prediction of patients with gastric cancer after neoadjuvant chemotherapy is suboptimal. This study aims to develop and validate a dynamic radiomic model for prognosis prediction of patients with gastric cancer on the basis of baseline and posttreatment features.

PATIENTS AND METHODS

This single-center cohort study included patients with gastric adenocarcinoma treated with neoadjuvant chemotherapy from June 2009 to July 2015 in the Gastrointestinal Cancer Center of Peking University Cancer Hospital. Their clinicopathological data, pre-treatment and post-treatment computed tomography (CT) images, and pathological reports were retrieved and analyzed. Four prediction models were developed and validated using tenfold cross-validation, with death within 3 years as the outcome. Model discrimination was compared by the area under the curve (AUC). The final radiomic model was evaluated for calibration and clinical utility using Hosmer-Lemeshow tests and decision curve analysis.

RESULTS

The study included 205 patients with gastric adenocarcinoma [166 (81%) male; mean age 59.9 (SD 10.3) years], with 71 (34.6%) deaths occurring within 3 years. The radiomic model alone demonstrated better discrimination than the pathological T stage (ypT) stage model alone (cross-validated AUC 0.598 versus 0.516, P = 0.009). The final radiomic model, which incorporated both radiomic and clinicopathological characteristics, had a significantly higher cross-validated AUC (0.769) than the ypT stage model (0.516), the radiomics alone model (0.598), and the ypT plus other clinicopathological characteristics model (0.738; all P < 0.05). Decision curve analysis confirmed the clinical utility of the final radiomic model.

CONCLUSIONS

The developed radiomic model had good accuracy and could be used as a decision aid tool in clinical practice to differentiate prognosis of patients with gastric cancer.

Artificial Intelligence Modeling to Predict Periprosthetic Infection and Explantation following Implant-Based Reconstruction

BACKGROUND

Despite improvements in prosthesis design and surgical techniques, periprosthetic infection and explantation rates following implant-based reconstruction (IBR) remain relatively high. Artificial intelligence is an extremely powerful predictive tool that involves machine learning (ML) algorithms. We sought to develop, validate, and evaluate the use of ML algorithms to predict complications of IBR.

METHODS

A comprehensive review of patients who underwent IBR from January of 2018 to December of 2019 was conducted. Nine supervised ML algorithms were developed to predict periprosthetic infection and explantation. Patient data were randomly divided into training (80%) and testing (20%) sets.

RESULTS

The authors identified 481 patients (694 reconstructions) with a mean ± SD age of 50.0 ± 11.5 years, mean ± SD body mass index of 26.7 ± 4.8 kg/m 2 , and median follow-up time of 16.1 months (range, 11.9 to 3.2 months). Periprosthetic infection developed in 113 of the reconstructions (16.3%), and explantation was required with 82 (11.8%) of them. ML demonstrated good discriminatory performance in predicting periprosthetic infection and explantation (area under the receiver operating characteristic curve, 0.73 and 0.78, respectively), and identified nine and 12 significant predictors of periprosthetic infection and explantation, respectively.

CONCLUSIONS

ML algorithms trained using readily available perioperative clinical data accurately predict periprosthetic infection and explantation following IBR. The authors' findings support incorporating ML models into perioperative assessment of patients undergoing IBR to provide data-driven, patient-specific risk assessment to aid individualized patient counseling, shared decision-making, and presurgical optimization.

Derivation and validation of risk prediction for posttraumatic stress symptoms following trauma exposure

BACKGROUND

Posttraumatic stress symptoms (PTSS) are common following traumatic stress exposure (TSE). Identification of individuals with PTSS risk in the early aftermath of TSE is important to enable targeted administration of preventive interventions. In this study, we used baseline survey data from two prospective cohort studies to identify the most influential predictors of substantial PTSS.

METHODS

Self-identifying black and white American women and men (n = 1546) presenting to one of 16 emergency departments (EDs) within 24 h of motor vehicle collision (MVC) TSE were enrolled. Individuals with substantial PTSS (⩾33, Impact of Events Scale - Revised) 6 months after MVC were identified via follow-up questionnaire. Sociodemographic, pain, general health, event, and psychological/cognitive characteristics were collected in the ED and used in prediction modeling. Ensemble learning methods and Monte Carlo cross-validation were used for feature selection and to determine prediction accuracy. External validation was performed on a hold-out sample (30% of total sample).

RESULTS

Twenty-five percent (n = 394) of individuals reported PTSS 6 months following MVC. Regularized linear regression was the top performing learning method. The top 30 factors together showed good reliability in predicting PTSS in the external sample (Area under the curve = 0.79 ± 0.002). Top predictors included acute pain severity, recovery expectations, socioeconomic status, self-reported race, and psychological symptoms.

CONCLUSIONS

These analyses add to a growing literature indicating that influential predictors of PTSS can be identified and risk for future PTSS estimated from characteristics easily available/assessable at the time of ED presentation following TSE.

Development and Assessment of Machine Learning Models for Individualized Risk Assessment of Mastectomy Skin Flap Necrosis

OBJECTIVE

To develop, validate, and evaluate ML algorithms for predicting MSFN.

BACKGROUND

MSFN is a devastating complication that causes significant distress to patients and physicians by prolonging recovery time, compromising surgical outcomes, and delaying adjuvant therapy.

METHODS

We conducted comprehensive review of all consecutive patients who underwent mastectomy and immediate implant-based reconstruction from January 2018 to December 2019. Nine supervised ML algorithms were developed to predict MSFN. Patient data were partitioned into training (80%) and testing (20%) sets.

RESULTS

We identified 694 mastectomies with immediate implant-based reconstruction in 481 patients. The patients had a mean age of 50 ± 11.5 years, years, a mean body mass index of 26.7 ± 4.8 kg/m 2 , and a median follow-up time of 16.1 (range, 11.9-23.2) months. MSFN developed in 6% (n = 40) of patients. The random forest model demonstrated the best discriminatory performance (area under curve, 0.70), achieved a mean accuracy of 89% (95% confidence interval, 83-94), and identified 10 predictors of MSFN. Decision curve analysis demonstrated that ML models have a superior net benefit regardless of the probability threshold. Higher body mass index, older age, hypertension, subpectoral device placement, nipple-sparing mastectomy, axillary nodal dissection, and no acellular dermal matrix use were all independently associated with a higher risk of MSFN.

CONCLUSIONS

ML algorithms trained on readily available perioperative clinical data can accurately predict the occurrence of MSFN and aid in individualized patient counseling, preoperative optimization, and surgical planning to reduce the risk of this devastating complication.

Artificial intelligence in theranostics of gastric cancer, a review

Gastric cancer (GC) is one of the commonest cancers with high morbidity and mortality in the world. How to realize precise diagnosis and therapy of GC owns great clinical requirement. In recent years, artificial intelligence (AI) has been actively explored to apply to early diagnosis and treatment and prognosis of gastric carcinoma. Herein, we review recent advance of AI in early screening, diagnosis, therapy and prognosis of stomach carcinoma. Especially AI combined with breath screening early GC system improved 97.4 % of early GC diagnosis ratio, AI model on stomach cancer diagnosis system of saliva biomarkers obtained an overall accuracy of 97.18 %, specificity of 97.44 %, and sensitivity of 96.88 %. We also discuss concept, issues, approaches and challenges of AI applied in stomach cancer. This review provides a comprehensive view and roadmap for readers working in this field, with the aim of pushing application of AI in theranostics of stomach cancer to increase the early discovery ratio and curative ratio of GC patients.

Evaluation of stacked ensemble model performance to predict clinical outcomes: A COVID-19 study

BACKGROUND

The application of machine learning (ML) to analyze clinical data with the goal to predict patient outcomes has garnered increasing attention. Ensemble learning has been used in conjunction with ML to improve predictive performance. Although stacked generalization (stacking), a type of heterogeneous ensemble of ML models, has emerged in clinical data analysis, it remains unclear how to define the best model combinations for strong predictive performance. This study develops a methodology to evaluate the performance of "base" learner models and their optimized combination using "meta" learner models in stacked ensembles to accurately assess performance in the context of clinical outcomes.

METHODS

De-identified COVID-19 data was obtained from the University of Louisville Hospital, where a retrospective chart review was performed from March 2020 to November 2021. Three differently-sized subsets using features from the overall dataset were chosen to train and evaluate ensemble classification performance. The number of base learners chosen from several algorithm families coupled with a complementary meta learner was varied from a minimum of 2 to a maximum of 8. Predictive performance of these combinations was evaluated in terms of mortality and severe cardiac event outcomes using area-under-the-receiver-operating-characteristic (AUROC), F1, balanced accuracy, and kappa.

RESULTS

The results highlight the potential to accurately predict clinical outcomes, such as severe cardiac events with COVID-19, from routinely acquired in-hospital patient data. Meta learners Generalized Linear Model (GLM), Multi-Layer Perceptron (MLP), and Partial Least Squares (PLS) had the highest AUROC for both outcomes, while K-Nearest Neighbors (KNN) had the lowest. Performance trended lower in the training set as the number of features increased, and exhibited less variance in both training and validation across all feature subsets as the number of base learners increased.

CONCLUSION

This study offers a methodology to robustly evaluate ensemble ML performance when analyzing clinical data.

Ebenen der Explizierbarkeit für medizinische künstliche Intelligenz: Was brauchen wir normativ und was können wir technisch erreichen?

Definition of the problem

The umbrella term “explicability” refers to the reduction of opacity of artificial intelligence (AI) systems. These efforts are challenging for medical AI applications because higher accuracy often comes at the cost of increased opacity. This entails ethical tensions because physicians and patients desire to trace how results are produced without compromising the performance of AI systems. The centrality of explicability within the informed consent process for medical AI systems compels an ethical reflection on the trade-offs. Which levels of explicability are needed to obtain informed consent when utilizing medical AI?

Arguments

We proceed in five steps: First, we map the terms commonly associated with explicability as described in the ethics and computer science literature, i.e., disclosure, intelligibility, interpretability, and explainability. Second, we conduct a conceptual analysis of the ethical requirements for explicability when it comes to informed consent. Third, we distinguish hurdles for explicability in terms of epistemic and explanatory opacity. Fourth, this then allows to conclude the level of explicability physicians must reach and what patients can expect. In a final step, we show how the identified levels of explicability can technically be met from the perspective of computer science. Throughout our work, we take diagnostic AI systems in radiology as an example.

Conclusion

We determined four levels of explicability that need to be distinguished for ethically defensible informed consent processes and showed how developers of medical AI can technically meet these requirements.

A Deep-Learning Approach for Identifying and Classifying Digestive Diseases

The digestive tract, often known as the gastrointestinal (GI) tract or the gastrointestinal system, is affected by digestive ailments. The stomach, large and small intestines, liver, pancreas and gallbladder are all components of the digestive tract. A digestive disease is any illness that affects the digestive system. Serious to moderate conditions can exist. Heartburn, cancer, irritable bowel syndrome (IBS) and lactose intolerance are only a few of the frequent issues. The digestive system may be treated with many different surgical treatments. Laparoscopy, open surgery and endoscopy are a few examples of these techniques. This paper proposes transfer-learning models with different pre-trained models to identify and classify digestive diseases. The proposed systems showed an increase in metrics, such as the accuracy, precision and recall, when compared with other state-of-the-art methods, and EfficientNetB0 achieved the best performance results of 98.01% accuracy, 98% precision and 98% recall.

Automatic Detection of Oral Squamous Cell Carcinoma from Histopathological Images of Oral Mucosa Using Deep Convolutional Neural Network

Worldwide, oral cancer is the sixth most common type of cancer. India is in 2nd position, with the highest number of oral cancer patients. To the population of oral cancer patients, India contributes to almost one-third of the total count. Among several types of oral cancer, the most common and dominant one is oral squamous cell carcinoma (OSCC). The major reason for oral cancer is tobacco consumption, excessive alcohol consumption, unhygienic mouth condition, betel quid eating, viral infection (namely human papillomavirus), etc. The early detection of oral cancer type OSCC, in its preliminary stage, gives more chances for better treatment and proper therapy. In this paper, author proposes a convolutional neural network model, for the automatic and early detection of OSCC, and for experimental purposes, histopathological oral cancer images are considered. The proposed model is compared and analyzed with state-of-the-art deep learning models like VGG16, VGG19, Alexnet, ResNet50, ResNet101, Mobile Net and Inception Net. The proposed model achieved a cross-validation accuracy of 97.82%, which indicates the suitability of the proposed approach for the automatic classification of oral cancer data.

On the importance of interpretable machine learning predictions to inform clinical decision making in oncology

Machine learning-based tools are capable of guiding individualized clinical management and decision-making by providing predictions of a patient's future health state. Through their ability to model complex nonlinear relationships, ML algorithms can often outperform traditional statistical prediction approaches, but the use of nonlinear functions can mean that ML techniques may also be less interpretable than traditional statistical methodologies. While there are benefits of intrinsic interpretability, many model-agnostic approaches now exist and can provide insight into the way in which ML systems make decisions. In this paper, we describe how different algorithms can be interpreted and introduce some techniques for interpreting complex nonlinear algorithms.

A Surgeon's Guide to Artificial Intelligence-Driven Predictive Models

Artificial intelligence (AI) focuses on processing and interpreting complex information as well as identifying relationships and patterns among complex data. Artificial intelligence- and machine learning (ML)-driven predictions have shown promising potential in influencing real-time decisions and improving surgical outcomes by facilitating screening, diagnosis, risk assessment, preoperative planning, and shared decision-making. Fundamental understanding of the algorithms, as well as their development and interpretation, is essential for the evolution of AI in surgery. In this article, we provide surgeons with a fundamental understanding of AI-driven predictive models through an overview of common ML and deep learning algorithms, model development, performance metrics and interpretation. This would serve as a basis for understanding ML-based research, while fostering new ideas and innovations for furthering the reach of this emerging discipline.

Artificial Intelligence (AI) and Internet of Medical Things (IoMT) Assisted Biomedical Systems for Intelligent Healthcare

Artificial intelligence (AI) is a modern approach based on computer science that develops programs and algorithms to make devices intelligent and efficient for performing tasks that usually require skilled human intelligence. AI involves various subsets, including machine learning (ML), deep learning (DL), conventional neural networks, fuzzy logic, and speech recognition, with unique capabilities and functionalities that can improve the performances of modern medical sciences. Such intelligent systems simplify human intervention in clinical diagnosis, medical imaging, and decision-making ability. In the same era, the Internet of Medical Things (IoMT) emerges as a next-generation bio-analytical tool that combines network-linked biomedical devices with a software application for advancing human health. In this review, we discuss the importance of AI in improving the capabilities of IoMT and point-of-care (POC) devices used in advanced healthcare sectors such as cardiac measurement, cancer diagnosis, and diabetes management. The role of AI in supporting advanced robotic surgeries developed for advanced biomedical applications is also discussed in this article. The position and importance of AI in improving the functionality, detection accuracy, decision-making ability of IoMT devices, and evaluation of associated risks assessment is discussed carefully and critically in this review. This review also encompasses the technological and engineering challenges and prospects for AI-based cloud-integrated personalized IoMT devices for designing efficient POC biomedical systems suitable for next-generation intelligent healthcare.

Artificial intelligence and machine learning in precision and genomic medicine

The advancement of precision medicine in medical care has led behind the conventional symptom-driven treatment process by allowing early risk prediction of disease through improved diagnostics and customization of more effective treatments. It is necessary to scrutinize overall patient data alongside broad factors to observe and differentiate between ill and relatively healthy people to take the most appropriate path toward precision medicine, resulting in an improved vision of biological indicators that can signal health changes. Precision and genomic medicine combined with artificial intelligence have the potential to improve patient healthcare. Patients with less common therapeutic responses or unique healthcare demands are using genomic medicine technologies. AI provides insights through advanced computation and inference, enabling the system to reason and learn while enhancing physician decision making. Many cell characteristics, including gene up-regulation, proteins binding to nucleic acids, and splicing, can be measured at high throughput and used as training objectives for predictive models. Researchers can create a new era of effective genomic medicine with the improved availability of a broad range of datasets and modern computer techniques such as machine learning. This review article has elucidated the contributions of ML algorithms in precision and genome medicine.

Artificial intelligence and machine learning in precision and genomic medicine

The advancement of precision medicine in medical care has led behind the conventional symptom-driven treatment process by allowing early risk prediction of disease through improved diagnostics and customization of more effective treatments. It is necessary to scrutinize overall patient data alongside broad factors to observe and differentiate between ill and relatively healthy people to take the most appropriate path toward precision medicine, resulting in an improved vision of biological indicators that can signal health changes. Precision and genomic medicine combined with artificial intelligence have the potential to improve patient healthcare. Patients with less common therapeutic responses or unique healthcare demands are using genomic medicine technologies. AI provides insights through advanced computation and inference, enabling the system to reason and learn while enhancing physician decision making. Many cell characteristics, including gene up-regulation, proteins binding to nucleic acids, and splicing, can be measured at high throughput and used as training objectives for predictive models. Researchers can create a new era of effective genomic medicine with the improved availability of a broad range of datasets and modern computer techniques such as machine learning. This review article has elucidated the contributions of ML algorithms in precision and genome medicine.

Comparison of machine learning and deep learning for view identification from cardiac magnetic resonance images

BACKGROUND

Artificial intelligence is increasingly utilized to aid in the interpretation of cardiac magnetic resonance (CMR) studies. One of the first steps is the identification of the imaging plane depicted, which can be achieved by both deep learning (DL) and classical machine learning (ML) techniques without user input. We aimed to compare the accuracy of ML and DL for CMR view classification and to identify potential pitfalls during training and testing of the algorithms.

METHODS

To train our DL and ML algorithms, we first established datasets by retrospectively selecting 200 CMR cases. The models were trained using two different cohorts (passively and actively curated) and applied data augmentation to enhance training. Once trained, the models were validated on an external dataset, consisting of 20 cases acquired at another center. We then compared accuracy metrics and applied class activation mapping (CAM) to visualize DL model performance.

RESULTS

The DL and ML models trained with the passively-curated CMR cohort were 99.1% and 99.3% accurate on the validation set, respectively. However, when tested on the CMR cases with complex anatomy, both models performed poorly. After training and testing our models again on all 200 cases (active cohort), validation on the external dataset resulted in 95% and 90% accuracy, respectively. The CAM analysis depicted heat maps that demonstrated the importance of carefully curating the datasets to be used for training.

CONCLUSIONS

Both DL and ML models can accurately classify CMR images, but DL outperformed ML when classifying images with complex heart anatomy.

Significance of Lymph Node Metastasis in the Treatment of Gastric Cancer and Current Challenges in Determining the Extent of Metastasis

The stomach exhibits abundant lymphatic flow, and metastasis to lymph nodes is common. In the case of gastric cancer, there is a regularity to the spread of lymph node metastasis, and it does not easily metastasize outside the regional nodes. Furthermore, when its extent is limited, nodal metastasis of gastric cancer can be cured by appropriate lymph node dissection. Therefore, identifying and determining the extent of lymph node metastasis is important for ensuring accurate diagnosis and appropriate surgical treatment in patients with gastric cancer. However, precise detection of lymph node metastasis remains difficult. Most nodal metastases in gastric cancer are microscopic metastases, which often occur in small-sized lymph nodes, and are thus difficult to diagnose both preoperatively and intraoperatively. Preoperative nodal diagnoses are mainly made using computed tomography, although the specificity of this method is low because it is mainly based on the size of the lymph node. Furthermore, peripheral nodal metastases cannot be palpated intraoperatively, nodal harvesting of resected specimens remains difficult, and the number of lymph nodes detected vary greatly depending on the skill of the technician. Based on these findings, gastrectomy with prophylactic lymph node dissection is considered the standard surgical procedure for gastric cancer. In contrast, several groups have examined the value of sentinel node biopsy for accurately evaluating nodal metastasis in patients with early gastric cancer, reporting high sensitivity and accuracy. Sentinel node biopsy is also important for individualizing and optimizing the extent of uniform prophylactic lymph node dissection and determining whether patients are indicated for function-preserving curative gastrectomy, which is superior in preventing post-gastrectomy symptoms and maintaining dietary habits. Notably, advancements in surgical treatment for early gastric cancer are expected to result in individualized surgical strategies with sentinel node biopsy. Chemotherapy for advanced gastric cancer has also progressed, and conversion gastrectomy can now be performed after downstaging, even in cases previously regarded as inoperable. In this review, we discuss the importance of determining lymph node metastasis in the treatment of gastric cancer, the associated difficulties, and the need to investigate strategies that can improve the diagnosis of lymph node metastasis.

Differentiation of Jugular Foramen Paragangliomas versus Schwannomas Using Golden-Angle Radial Sparse Parallel Dynamic Contrast-Enhanced MRI

BACKGROUND AND PURPOSE

Accurate differentiation of paragangliomas and schwannomas in the jugular foramen has important clinical implications because treatment strategies may vary but differentiation is not always straightforward with conventional imaging. Our aim was to evaluate the accuracy of both qualitative and quantitative metrics derived from dynamic contrast-enhanced MR imaging using golden-angle radial sparse parallel MR imaging to differentiate paragangliomas and schwannomas in the jugular foramen.

MATERIALS AND METHODS

A retrospective study of imaging data was performed on patients (n = 30) undergoing MR imaging for jugular foramen masses with the golden-angle radial sparse parallel MR imaging technique. Imaging data were postprocessed to obtain time-intensity curves and quantitative parameters. Data were normalized to the dural venous sinus for relevant parameters and analyzed for statistical significance using a Student t test. A univariate logistic model was created with a binary output, paraganglioma or schwannoma, using a wash-in rate as a variable. Additionally, lesions were clustered on the basis of the wash-in rate and washout rate using a 3-nearest neighbors method.

RESULTS

There were 22 paragangliomas and 8 schwannomas. All paragangliomas demonstrated a type 3 time-intensity curve, and all schwannomas demonstrated a type 1 time-intensity curve. There was a statistically significant difference between paragangliomas and schwannomas when comparing their values for area under the curve, peak enhancement, wash-in rate, and washout rate. A univariate logistic model with a binary output (paraganglioma or schwannoma) using wash-in rate as a variable was able to correctly predict all observed lesions (P < .001). All 30 lesions were classified correctly by using a 3-nearest neighbors method.

CONCLUSIONS

Paragangliomas at the jugular foramen can be reliably differentiated from schwannomas using golden-angle radial sparse parallel MR imaging-dynamic contrast-enhanced imaging when imaging characteristics cannot suffice.

Multi-institutional development and external validation of machine learning-based models to predict relapse risk of pancreatic ductal adenocarcinoma after radical resection

BACKGROUND

Surgical resection is the only potentially curative treatment for pancreatic ductal adenocarcinoma (PDAC) and the survival of patients after radical resection is closely related to relapse. We aimed to develop models to predict the risk of relapse using machine learning methods based on multiple clinical parameters.

METHODS

Data were collected and analysed of 262 PDAC patients who underwent radical resection at 3 institutions between 2013 and 2017, with 183 from one institution as a training set, 79 from the other 2 institution as a validation set. We developed and compared several predictive models to predict 1- and 2-year relapse risk using machine learning approaches.

RESULTS

Machine learning techniques were superior to conventional regression-based analyses in predicting risk of relapse of PDAC after radical resection. Among them, the random forest (RF) outperformed other methods in the training set. The highest accuracy and area under the receiver operating characteristic curve (AUROC) for predicting 1-year relapse risk with RF were 78.4% and 0.834, respectively, and for 2-year relapse risk were 95.1% and 0.998. However, the support vector machine (SVM) model showed better performance than the others for predicting 1-year relapse risk in the validation set. And the k neighbor algorithm (KNN) model achieved the highest accuracy and AUROC for predicting 2-year relapse risk.

CONCLUSIONS

By machine learning, this study has developed and validated comprehensive models integrating clinicopathological characteristics to predict the relapse risk of PDAC after radical resection which will guide the development of personalized surveillance programs after surgery.

Gated Graph Attention Network for Cancer Prediction

With its increasing incidence, cancer has become one of the main causes of worldwide mortality. In this work, we mainly propose a novel attention-based neural network model named Gated Graph ATtention network (GGAT) for cancer prediction, where a gating mechanism (GM) is introduced to work with the attention mechanism (AM), to break through the previous work's limitation of 1-hop neighbourhood reasoning. In this way, our GGAT is capable of fully mining the potential correlation between related samples, helping for improving the cancer prediction accuracy. Additionally, to simplify the datasets, we propose a hybrid feature selection algorithm to strictly select gene features, which significantly reduces training time without affecting prediction accuracy. To the best of our knowledge, our proposed GGAT achieves the state-of-the-art results in cancer prediction task on LIHC, LUAD, KIRC compared to other traditional machine learning methods and neural network models, and improves the accuracy by 1% to 2% on Cora dataset, compared to the state-of-the-art graph neural network methods.

Artificial intelligence performance in detecting tumor metastasis from medical radiology imaging: A systematic review and meta-analysis

BACKGROUND

Early diagnosis of tumor metastasis is crucial for clinical treatment. Artificial intelligence (AI) has shown great promise in the field of medicine. We therefore aimed to evaluate the diagnostic accuracy of AI algorithms in detecting tumor metastasis using medical radiology imaging.

METHODS

We searched PubMed and Web of Science for studies published from January 1, 1997, to January 30, 2020. Studies evaluating an AI model for the diagnosis of tumor metastasis from medical images were included. We excluded studies that used histopathology images or medical wave-form data and those focused on the region segmentation of interest. Studies providing enough information to construct contingency tables were included in a meta-analysis.

FINDINGS

We identified 2620 studies, of which 69 were included. Among them, 34 studies were included in a meta-analysis with a pooled sensitivity of 82% (95% CI 79-84%), specificity of 84% (82-87%) and AUC of 0·90 (0·87-0·92). Analysis for different AI algorithms showed a pooled sensitivity of 87% (83-90%) for machine learning and 86% (82-89%) for deep learning, and a pooled specificity of 89% (82-93%) for machine learning, and 87% (82-91%) for deep learning.

INTERPRETATION

AI algorithms may be used for the diagnosis of tumor metastasis using medical radiology imaging with equivalent or even better performance to health-care professionals, in terms of sensitivity and specificity. At the same time, rigorous reporting standards with external validation and comparison to health-care professionals are urgently needed for AI application in the medical field.

FUNDING

College students' innovative entrepreneurial training plan program .

Artificial intelligence in gastric cancer: Application and future perspectives

Gastric cancer is the fourth leading cause of cancer-related mortality across the globe, with a 5-year survival rate of less than 40%. In recent years, several applications of artificial intelligence (AI) have emerged in the gastric cancer field based on its efficient computational power and learning capacities, such as image-based diagnosis and prognosis prediction. AI-assisted diagnosis includes pathology, endoscopy, and computerized tomography, while researchers in the prognosis circle focus on recurrence, metastasis, and survival prediction. In this review, a comprehensive literature search was performed on articles published up to April 2020 from the databases of PubMed, Embase, Web of Science, and the Cochrane Library. Thereby the current status of AI-applications was systematically summarized in gastric cancer. Moreover, future directions that target this field were also analyzed to overcome the risk of overfitting AI models and enhance their accuracy as well as the applicability in clinical practice.

Blood cancer diagnosis using ensemble learning based on a random subspace method in laser-induced breakdown spectroscopy

There are two main challenges in the diagnosis of blood cancer. The first is to diagnose cancer from healthy control, and the second is to identify the types of blood cancer. The chemometrics method combined with laser-induced breakdown spectroscopy (LIBS) can be used for cancer detection. However, chemometrics methods were easily influenced by the spectral feature redundancy and noise, resulting in low accuracy rate because of their simple structure. We proposed an approach using LIBS combined with the ensemble learning based on the random subspace method (RSM). The serum samples were dripped onto a boric acid substrate for LIBS spectrum collection. The complete blood cancer sample set include leukemia [acute myeloid leukemia (AML) and chronic myelogenous leukemia (CML)], multiple myeloma (MM), and lymphoma. The results showed that the accuracy rates using k nearest neighbors (kNN) and linear discriminant analysis (LDA) only were 88.14% and 94.45%, respectively, while using RSM with LDA (RSM-LDA), the average accuracy rate was improved from 94.45% to 98.34%. Furthermore, the variable importance of spectral lines (Na, K, Mg, Ca, H, O, N, C-N) were evaluated by the RSM-LDA model, which can improve the recognition ability of blood cancer types. Comparing the RSM-LDA model and only with LDA, the results showed that the average accuracy rate for cancer type identification was improved from 80.4% to 91.0%. These results demonstrate that LIBS combined with the RSM-LDA model can discriminate the blood cancer from the health control, as well as the recognition the types for blood cancers.

The major effects of health-related quality of life on 5-year survival prediction among lung cancer survivors: applications of machine learning

The primary goal of this study was to evaluate the major roles of health-related quality of life (HRQOL) in a 5-year lung cancer survival prediction model using machine learning techniques (MLTs). The predictive performances of the models were compared with data from 809 survivors who underwent lung cancer surgery. Each of the modeling technique was applied to two feature sets: feature set 1 included clinical and sociodemographic variables, and feature set 2 added HRQOL factors to the variables from feature set 1. One of each developed prediction model was trained with the decision tree (DT), logistic regression (LR), bagging, random forest (RF), and adaptive boosting (AdaBoost) methods, and then, the best algorithm for modeling was determined. The models' performances were compared using fivefold cross-validation. For feature set 1, there were no significant differences in model accuracies (ranging from 0.647 to 0.713). Among the models in feature set 2, the AdaBoost and RF models outperformed the other prognostic models [area under the curve (AUC) = 0.850, 0.898, 0.981, 0.966, and 0.949 for the DT, LR, bagging, RF and AdaBoost models, respectively] in the test set. Overall, 5-year disease-free lung cancer survival prediction models with MLTs that included HRQOL as well as clinical variables improved predictive performance.

Integrating tumor and nodal radiomics to predict lymph node metastasis in gastric cancer

BACKGROUND

To develop and validate a radiomics method via integrating tumor and lymph node radiomics for the preoperative prediction of lymph node (LN) status in gastric cancer (GC).

MATERIALS AND METHODS

We retrospectively collected 170 contrast-enhanced abdominal CT images from GC patients. Five times repeated random hold-out experiment was employed. Tumor and nodal radiomics features were extracted from each individual tumor and LN respectively, and then multi-step feature selection was performed. The optimal tumor and nodal features were selected using Pearson correlation analysis and sequential forward floating selection (SFFS) algorithm. After feature fusion, the SFFS algorithm was used to develop radiomics signatures. The performance of the radiomics signatures developed based on logistic regression classifier was further analyzed and compared using the area under the receiver operating characteristic curve (AUC).

RESULTS

The AUC values, reported as mean ± standard deviation, were 0.9319 ± 0.0129 and 0.8546 ± 0.0261 for the training and validation cohorts respectively. The radiomic signatures could predict LN status, especially in T2-stage, diffuse-type and moderately/well differentiated GC. After integrating clinicopathologic information, the radiomic-clinicopathologic model (training cohort, 0.9432 ± 0.0129; validation cohort, 0.8764 ± 0.0322) showed a better discrimination capability than other radiomics models and clinicopathologic model. The radiomic-clinicopathologic model also showed superior performance to the gastroenterologist' decision in all experiments, and outperformed the radiologist in some experiments.

CONCLUSION

Our proposed method presented good predictive performance and great potential for predicting LNM in GC. As a noninvasive preoperative prediction tool, it can be helpful for guiding the prognosis and treatment decision-making in GC patients.

Predicting Frequency, Time-To-Repair and Costs of Wind Turbine Failures

Operation and maintenance (O&M) costs, and associated uncertainty, for wind turbines (WTs) is a significant burden for wind farm operators. Many wind turbine failures are unpredictable while causing loss of energy production, and may also cause loss of asset. This study utilized 753 O&M event data from 21 wind turbines operating in Germany, to improve the prediction of failure frequency and associated costs. We applied Bayesian updating to predict wind turbine failure frequency and time-to-repair (TTR), in conjunction to machine learning techniques for assessing costs associated with failures. We found that time-to-failure (TTF), time-to-repair and the cost of failures depend on operational and environmental conditions. High elevation (>100 m) of the wind turbine installation was found to increase both the probability of failures and probability of delayed repairs. Furthermore, it was determined that direct-drive turbines are more favorable at locations with high capacity factor (more than 40%) whereas geared-drive turbines show lower failure costs than direct-drive ones at temperate-coastal locations with medium capacity factors (between 20% and 40%). Based on these findings, we developed a decision support tool that can guide a site-specific selection of wind turbine types, while providing a thorough estimation of O&M budgets.

Resectable pancreatic ductal adenocarcinoma: association between preoperative CT texture features and metastatic nodal involvement

Background

To explore the relationship between the lymph node status and preoperative computed tomography images texture features in pancreatic cancer.

Methods

A total of 155 operable pancreatic cancer patients (104 men, 51 women; mean age 63.8 ± 9.6 years), who had undergone contrast-enhanced computed tomography in the arterial and portal venous phases, were enrolled in this retrospective study. There were 73 patients with lymph node metastases and 82 patients without nodal involvement. Four different data sets, with thin (1.25 mm) and thick (5 mm) slices (at arterial phase and portal venous phase) were analysed. Texture analysis was performed by using MaZda software. A combination of feature selection algorithms was used to determine 30 texture features with the optimal discriminative performance for differentiation between lymph node positive and negative groups. The prediction performance of the selected feature was evaluated by receiver operating characteristic (ROC) curve analysis.

Results

There were 10 texture features with significant differences between two groups and significance in ROC analysis were identified. They were WavEnLH_s-2(wavelet energy with rows and columns are filtered with low pass and high pass frequency bands with scale factors 2) from wavelet-based features, 135dr_LngREmph (long run emphasis in 135 direction) and 135dr_Fraction (fraction of image in runs in 135 direction) from run length matrix-based features, and seven variables of sum average from coocurrence matrix-based features (SumAverg). The ideal cutoff value for predicting lymph node metastases was 270 for WavEnLH_s-2 (positive likelihood ratio 2.08). In addition, 135dr_LngREmph and 135dr_Fraction were correlated with the ratio of metastatic to examined lymph nodes.

Conclusions

Preoperative computed tomography high order texture features provide a useful imaging signature for the prediction of nodal involvement in pancreatic cancer.

Artificial intelligence with multi-functional machine learning platform development for better healthcare and precision medicine

Precision medicine is one of the recent and powerful developments in medical care, which has the potential to improve the traditional symptom-driven practice of medicine, allowing earlier interventions using advanced diagnostics and tailoring better and economically personalized treatments. Identifying the best pathway to personalized and population medicine involves the ability to analyze comprehensive patient information together with broader aspects to monitor and distinguish between sick and relatively healthy people, which will lead to a better understanding of biological indicators that can signal shifts in health. While the complexities of disease at the individual level have made it difficult to utilize healthcare information in clinical decision-making, some of the existing constraints have been greatly minimized by technological advancements. To implement effective precision medicine with enhanced ability to positively impact patient outcomes and provide real-time decision support, it is important to harness the power of electronic health records by integrating disparate data sources and discovering patient-specific patterns of disease progression. Useful analytic tools, technologies, databases, and approaches are required to augment networking and interoperability of clinical, laboratory and public health systems, as well as addressing ethical and social issues related to the privacy and protection of healthcare data with effective balance. Developing multifunctional machine learning platforms for clinical data extraction, aggregation, management and analysis can support clinicians by efficiently stratifying subjects to understand specific scenarios and optimize decision-making. Implementation of artificial intelligence in healthcare is a compelling vision that has the potential in leading to the significant improvements for achieving the goals of providing real-time, better personalized and population medicine at lower costs. In this study, we focused on analyzing and discussing various published artificial intelligence and machine learning solutions, approaches and perspectives, aiming to advance academic solutions in paving the way for a new data-centric era of discovery in healthcare.

Detection of gastric cancer and its histological type based on iodine concentration in spectral CT

Background

Computed tomography (CT) imaging is the most common imaging modality for the diagnosis and staging of gastric cancer. The aim of this study is was to prospectively explore the ability of quantitative spectral CT parameters in the detection of gastric cancer and its histologic types.

Methods

A total of 87 gastric adenocarcinoma (43 poorly and 44 well-differentiated) patients and 36 patients with benign gastric wall lesions (25 inflammation and 11 normal), who underwent dual-phase enhanced spectral CT examination, were retrospectively enrolled in this study. Iodine concentration (IC) and normalized iodine concentration (nIC) during arterial phase (AP) and portal venous phase (PP) were measured thrice in each patient by two blinded radiologists. Moreover, intraclass correlation coefficient (ICC) was used to assess the interobserver reproducibility. Differences of IC and nIC values between gastric cancer and benign lesion groups were compared using Mann-Whitney U test. Furthermore, the gender, age, location, thickness and histological types of gastric adenocarcinoma were analyzed by Mann-Whitney U test or Kruskal-Wallis H test. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic efficacy of IC and nIC values, and the optimal cut-off value was calculated with Youden J.

Results

An excellent interobserver agreement (ICC >  0.6) was achieved for IC. Notably, the values of ICAP, ICPP, nICAP and nICPP were significantly higher in gastric cancer group (Z = 5.870, 3.894, 2.009 and 10.137, respectively; P < 0.05) than those in benign lesion group. Additionally, the values of ICAP, ICPP, nICAP and nICPP were significantly higher in poorly differentiated gastric adenocarcinoma group (Z = 4.118, 5.637, 6.729 and 2.950, respectively; P < 0.005) than those in well-differentiated gastric adenocarcinoma group. There were no statistically significant differences in the values of ICAP, ICPP, nICAP and nICPP between age, gender, tumor thickness and tumor location. Furthermore, the area under the curve (AUC) values of ICAP, nICAP, ICPP and nICPP were 0.745, 0.584, 0.662, and 0.932, respectively, for gastric cancer detection; while 0.756, 0.919, 0.851 and 0.684, respectively, in discriminating poorly differentiated gastric adenocarcinoma.

Conclusion

IC values exhibited great potential in the preoperative and non-invasive diagnosis of gastric cancer and its histological types. In particular, nICPP is more effective for the identification of gastric cancer, whereas nICAP is more effective in discriminating poorly differentiated gastric adenocarcinoma.

Liquid biopsies for hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is the world's second leading cause of cancer death; 82.4% of patients die within 5 years. This grim prognosis is the consequence of a lack of effective early detection tools, limited treatment options, and the high frequency of HCC recurrence. Advances in the field of liquid biopsy hold great promise in improving early detection of HCC, advancing patient prognosis, and ultimately increasing the survival rate. In an effort to address the current challenges of HCC screening and management, several studies have identified and evaluated liver-cancer-associated molecular signatures such as genetic alterations, methylation, and noncoding RNA expression in the form of circulating biomarkers in body fluids and circulating tumor cells of HCC patients. In this review, we summarize the recent progress in HCC liquid biopsy, organized by the intended clinical application of the reported study.

Detection of Sclerotinia Stem Rot on Oilseed Rape (Brassica napus L.) Leaves Using Hyperspectral Imaging

Hyperspectral imaging was explored to detect Sclerotinia stem rot (SSR) on oilseed rape leaves with chemometric methods, and the influences of variable selection, machine learning, and calibration transfer methods on detection performances were evaluated. Three different sample sets containing healthy and infected oilseed rape leaves were acquired under different imaging acquisition parameters. Four discriminant models were built using full spectra, including partial least squares-discriminant analysis (PLS-DA), support vector machine (SVM), soft independent modeling of class analogies (SIMCA), and k-nearest neighbors (KNN). PLS-DA and SVM models were also built with the optimal wavelengths selected by principal component analysis (PCA) loadings, second derivative spectra, competitive adaptive reweighted sampling (CARS), and successive projections algorithm (SPA). The optimal wavelengths selected for each sample set by different methods were different; however, the optimal wavelengths selected by PCA loadings and second derivative spectra showed similarity between different sample sets. Direct standardization (DS) was successfully applied to reduce spectral differences among different sample sets. Overall, the results demonstrated that using hyperspectral imaging with chemometrics for plant disease detection can be efficient and will also help in the selection of optimal variable selection, machine learning, and calibration transfer methods for fast and accurate plant disease detection.

Model-based quantitative optical biopsy in multilayer in vitro soft tissue models for whole field assessment of nonmelanoma skin cancer

Optical techniques such as fluorescence and diffuse reflectance spectroscopy are proven to have the potential to provide tissue discrimination during the development of malignancies and hence treated as potential tools for noninvasive optical biopsy in clinical diagnostics. Quantitative optical biopsy is challenging and hence the majority of the existing strategies are based on a qualitative assessment of the concerned tissue. Light-tissue interaction models as well as precise optical phantoms can greatly help in the former and here we present a pilot study to assess the optical properties of a multilayer tissue-specific optical phantom with the help of a database generated using multilayer-Monte Carlo (MCML) models. A set of optical models mimicking the properties of actual and diseased conditions of tissues associated with nonmelanoma skin cancer (NMSC) were devised and MCML simulations of fluorescence and diffuse reflectance were performed on these models to generate the spectral signature of identified biomarkers of NMSC such as hemoglobin, flavin adenine dinucleotide, and collagen. A model library was generated and with the extracted features from modeled spectra, classification of normal and NMSC conditions were tested using the [Formula: see text]-nearest neighbor (KNN) classifier. Using an in-house assembled scan-based automated bimodal spectral imaging system with reflectance and fluorescence modalities of operation, a layered, thin, tissue equivalent phantom, fabricated with controlled optical properties mimicking normal and NMSC conditions were tested. The spectral signatures corresponding to the NMSC biomarkers were acquired from this phantom and extracted features from the spectra were tested using the KNN classifier and classification accuracy of 100% was achieved. For further quantitative analysis, the experimental and simulated spectra were compared with respect to the light intensity at the emission peak or absorption dips, spectral line width, and average intensity over a range of wavelength of interest and observed to be analogous within specified and systematic error limits. This methodology is expected to give a better quantitative approach for estimation of tissue properties by correlating the experimental and simulated data.

Development and Evaluation of Novel Statistical Methods in Urine Biomarker-Based Hepatocellular Carcinoma Screening

Hepatocellular carcinoma is one of the fastest growing cancers in the US and has a low survival rate, partly due to difficulties in early detection. Because of HCC’s high heterogeneity, it has been suggested that multiple biomarkers would be needed to develop a sensitive HCC screening test. This study applied random forest (RF), a machine learning technique, and proposed two novel models, fixed sequential (FS) and two-step (TS), for comparison with two commonly used statistical techniques, logistic regression (LR) and classification and regression trees (CART), in combining multiple urine DNA biomarkers for HCC screening using biomarker values obtained from 137 HCC and 431 non-HCC (224 hepatitis and 207 cirrhosis) subjects. The sensitivity, specificity, area under the receiver operating curve, and variability were estimated through repeated 10-fold cross-validation to compare the models’ performances in accuracy and robustness. We show that RF and TS have higher accuracy and stability; specifically, they reach 90% specificity and 86%/87% sensitivity respectively along with 15% higher sensitivity and 10% higher specificity than LR in cross-validation. The potential of RF and TS to develop a panel of multiple biomarkers and the possibility for self-training, cloud-based models for HCC screening are discussed.

Advances in imaging of gastric cancer

Due to the advances in imaging technology, the diagnosis, staging, and efficacy evaluation of gastric cancer by imaging are continuously improving. This paper comprehensively analyzes different imaging methods in cancer research, focusing on new imaging techniques for the diagnosis and treatment of gastric cancer, their advantages and limitations in clinical application, and the opportunities and challenges. Radiologists can take the initiative to collaborate with relevant clinical departments through a multidisciplinary platform with an open mind in the face of various problems presented clinically, understand the requirements for standardized diagnosis and treatment of gastric cancer, and fully communicate with imaging equipment providers and engineering and technical personnel to explore more methods and indicators to improve the diagnosis of this malignancy.

Application of computed tomography virtual noncontrast spectral imaging in evaluation of hepatic metastases: a preliminary study

Objective:

The objective was to qualitatively and quantitatively evaluate hepatic metastases using computed tomography (CT) virtual noncontrast (VNC) spectral imaging in a retrospective analysis.

Methods:

Forty hepatic metastases patients underwent CT scans including the conventional true noncontrast (TNC) and the tri-phasic contrast-enhanced dual energy spectral scans in the hepatic arterial, portal venous, and equilibrium phases. The tri-phasic spectral CT images were used to obtain three groups of VNC images including in the arterial (VNCa), venous (VNCv), and equilibrium (VNCe) phase by the material decomposition process using water and iodine as a base material pair. The image quality and the contrast-to-noise ratio (CNR) of metastasis of the four groups were compared with ANOVA analysis. The metastasis detection rates with the four nonenhanced image groups were calculated and compared using the Chi-square test.

Results:

There were no significant differences in image quality among TNC, VNCa and VNCv images (P > 0.05). The quality of VNCe images was significantly worse than that of other three groups (P < 0.05). The mean CNR of metastasis in the TNC and VNCs images was 1.86, 2.42, 1.92, and 1.94, respectively; the mean CNR of metastasis in VNCa images was significantly higher than that in other three groups (P < 0.05), while no statistically significant difference was observed among VNCv, VNCe and TNC images (P > 0.05). The metastasis detection rate of the four nonenhanced groups with no statistically significant difference (P > 0.05).

Conclusions:

The quality of VNCa and VNCv images is identical to that of TNC images, and the metastasis detection rate in VNC images is similar to that in TNC images. VNC images obtained from arterial phase show metastases more clearly. Thus, VNCa imaging may be a surrogate to TNC imaging in hepatic metastasis diagnosis.

Machine-learning prediction of cancer survival: a retrospective study using electronic administrative records and a cancer registry

OBJECTIVES

Using the prediction of cancer outcome as a model, we have tested the hypothesis that through analysing routinely collected digital data contained in an electronic administrative record (EAR), using machine-learning techniques, we could enhance conventional methods in predicting clinical outcomes.

SETTING

A regional cancer centre in Australia.

PARTICIPANTS

Disease-specific data from a purpose-built cancer registry (Evaluation of Cancer Outcomes (ECO)) from 869 patients were used to predict survival at 6, 12 and 24 months. The model was validated with data from a further 94 patients, and results compared to the assessment of five specialist oncologists. Machine-learning prediction using ECO data was compared with that using EAR and a model combining ECO and EAR data.

PRIMARY AND SECONDARY OUTCOME MEASURES

Survival prediction accuracy in terms of the area under the receiver operating characteristic curve (AUC).

RESULTS

The ECO model yielded AUCs of 0.87 (95% CI 0.848 to 0.890) at 6 months, 0.796 (95% CI 0.774 to 0.823) at 12 months and 0.764 (95% CI 0.737 to 0.789) at 24 months. Each was slightly better than the performance of the clinician panel. The model performed consistently across a range of cancers, including rare cancers. Combining ECO and EAR data yielded better prediction than the ECO-based model (AUCs ranging from 0.757 to 0.997 for 6 months, AUCs from 0.689 to 0.988 for 12 months and AUCs from 0.713 to 0.973 for 24 months). The best prediction was for genitourinary, head and neck, lung, skin, and upper gastrointestinal tumours.

CONCLUSIONS

Machine learning applied to information from a disease-specific (cancer) database and the EAR can be used to predict clinical outcomes. Importantly, the approach described made use of digital data that is already routinely collected but underexploited by clinical health systems.