Predicting total lung capacity from spirometry: a machine learning approach.

Background and objective: Spirometry patterns can suggest that a patient has a restrictive ventilatory impairment; however, lung volume measurements such as total lung capacity (TLC) are required to confirm the diagnosis. The aim of the study was to train a supervised machine learning model that can accurately estimate TLC values from spirometry and subsequently identify which patients would most benefit from undergoing a complete pulmonary function test. Methods: We trained three tree-based machine learning models on 51,761 spirometry data points with corresponding TLC measurements. We then compared model performance using an independent test set consisting of 1,402 patients. The best-performing model was used to retrospectively identify restrictive ventilatory impairment in the same test set. The algorithm was compared against different spirometry patterns commonly used to predict restriction. Results: The prevalence of restrictive ventilatory impairment in the test set is 16.7% (234/1402). CatBoost was the best-performing machine learning model. It predicted TLC with a mean squared error (MSE) of 560.1 mL. The sensitivity, specificity, and F1-score of the optimal algorithm for predicting restrictive ventilatory impairment was 83, 92, and 75%, respectively. Conclusion: A machine learning model trained on spirometry data can estimate TLC to a high degree of accuracy. This approach could be used to develop future smart home-based spirometry solutions, which could aid decision making and self-monitoring in patients with restrictive lung diseases.

Artificial intelligence and machine learning for medical imaging: A technology review.

Artificial intelligence (AI) has recently become a very popular buzzword, as a consequence of disruptive technical advances and impressive experimental results, notably in the field of image analysis and processing. In medicine, specialties where images are central, like radiology, pathology or oncology, have seized the opportunity and considerable efforts in research and development have been deployed to transfer the potential of AI to clinical applications. With AI becoming a more mainstream tool for typical medical imaging analysis tasks, such as diagnosis, segmentation, or classification, the key for a safe and efficient use of clinical AI applications relies, in part, on informed practitioners. The aim of this review is to present the basic technological pillars of AI, together with the state-of-the-art machine learning methods and their application to medical imaging. In addition, we discuss the new trends and future research directions. This will help the reader to understand how AI methods are now becoming an ubiquitous tool in any medical image analysis workflow and pave the way for the clinical implementation of AI-based solutions.

Reinforcement Learning for Radiotherapy Dose Fractioning Automation.

External beam radiotherapy cancer treatment aims to deliver dose fractions to slowly destroy a tumor while avoiding severe side effects in surrounding healthy tissues. To automate the dose fraction schedules, this paper investigates how deep reinforcement learning approaches (based on deep Q network and deep deterministic policy gradient) can learn from a model of a mixture of tumor and healthy cells. A 2D tumor growth simulation is used to simulate radiation effects on tissues and thus training an agent to automatically optimize dose fractionation. Results show that initiating treatment with large dose per fraction, and then gradually reducing it, is preferred to the standard approach of using a constant dose per fraction.

Correction to: Tumor fragmentation estimated by volume surface ratio of tumors measured on 18F-FDG PET/CT is an independent prognostic factor of diffuse large B-cell lymphoma.

A unit error concerning the tumor volume surface ratio (TVSR) is present throughout the article. The unit reported is "cm" but is actually "mm".

Tumor fragmentation estimated by volume surface ratio of tumors measured on 18F-FDG PET/CT is an independent prognostic factor of diffuse large B-cell lymphoma.

INTRODUCTION: Our aim was to study the prognostic value of two new 18F-FDG PET biomarkers in diffuse large B-cell lymphoma (DLBCL). We examined the total tumor surface (TTS), describing the tumor-host interface, and the tumor volume surface ratio (TVSR), corresponding to the ratio between the total metabolic tumor volume (TMTV) and TTS, describing the tumor fragmentation. METHODS: We retrospectively included 215 patients with DLBCL. Patients underwent initial 18F-FDG PET/CT before R-CHOP (73%) or intensified R-CHOP (R-ACVBP) regimens (27%). The TMTV was measured using a fixed threshold value of 41% of SUVmax. To calculate TTS and TVSR, the surface was measured using an in-house software based on the marching cube algorithm. Spearman's rank correlation coefficient (ρ) was computed between TMTV, TTS, and TVSR, and ROC analysis was performed. Survival functions at 5 years were studied using a Kaplan-Meier method and uni/multivariate Cox analysis. RESULTS: TVSR was poorly correlated with TMTV (ρ = 0.5) and TTS (ρ = 0.26), while TTS was highly correlated with TMTV (ρ = 0.94) and was, therefore, excluded from the analysis. TMTV had the highest area under the ROC curve (0.711) and the best sensitivity (0.797), while TVSR had the best specificity (0.745). The optimal cut-off values to predict 5-year OS were 222 cm3 for TMTV and 6.0 mm for TVSR. Patients with high TMTV and TVSR had significantly worse prognosis in Kaplan-Meier and Cox univariate analysis. In a multivariate Cox analysis combining the International Prognostic Index (IPI), the type of chemotherapy, TMTV, and TVSR, all parameters were independent and significant prognostic factors (HR [95%CI]: IPI 1.4 [1.1-1.8], type of chemotherapy 4.5 [2.0-10.5], TMTV 2.8 [1.4-5.5], TVSR 2.1 [1.3-3.4]). A synergistic effect between TMTV and TVSR was observed in a Kaplan-Meier analysis combining the two parameters. CONCLUSIONS: TVSR measured on the initial 18F-FDG PET is an independent prognostic factor in DLBCL and has an additional prognostic value when combined with TMTV, IPI score and chemotherapy.

Predictive value of initial FDG-PET features for treatment response and survival in esophageal cancer patients treated with chemo-radiation therapy using a random forest classifier.

PURPOSE: In oncology, texture features extracted from positron emission tomography with 18-fluorodeoxyglucose images (FDG-PET) are of increasing interest for predictive and prognostic studies, leading to several tens of features per tumor. To select the best features, the use of a random forest (RF) classifier was investigated. METHODS: Sixty-five patients with an esophageal cancer treated with a combined chemo-radiation therapy were retrospectively included. All patients underwent a pretreatment whole-body FDG-PET. The patients were followed for 3 years after the end of the treatment. The response assessment was performed 1 month after the end of the therapy. Patients were classified as complete responders and non-complete responders. Sixty-one features were extracted from medical records and PET images. First, Spearman's analysis was performed to eliminate correlated features. Then, the best predictive and prognostic subsets of features were selected using a RF algorithm. These results were compared to those obtained by a Mann-Whitney U test (predictive study) and a univariate Kaplan-Meier analysis (prognostic study). RESULTS: Among the 61 initial features, 28 were not correlated. From these 28 features, the best subset of complementary features found using the RF classifier to predict response was composed of 2 features: metabolic tumor volume (MTV) and homogeneity from the co-occurrence matrix. The corresponding predictive value (AUC = 0.836 ± 0.105, Se = 82 ± 9%, Sp = 91 ± 12%) was higher than the best predictive results found using the Mann-Whitney test: busyness from the gray level difference matrix (P < 0.0001, AUC = 0.810, Se = 66%, Sp = 88%). The best prognostic subset found using RF was composed of 3 features: MTV and 2 clinical features (WHO status and nutritional risk index) (AUC = 0.822 ± 0.059, Se = 79 ± 9%, Sp = 95 ± 6%), while no feature was significantly prognostic according to the Kaplan-Meier analysis. CONCLUSIONS: The RF classifier can improve predictive and prognostic values compared to the Mann-Whitney U test and the univariate Kaplan-Meier survival analysis when applied to several tens of features in a limited patient database.