Is radiomics a useful addition to magnetic resonance imaging in the preoperative classification of PitNETs?

BACKGROUND: The WHO 2021 introduced the term pituitary neuroendocrine tumours (PitNETs) for pituitary adenomas and incorporated transcription factors for subtyping, prompting the need for fresh diagnostic methods. Current biomarkers struggle to distinguish between high- and low-risk non-functioning PitNETs. We explored if radiomics can enhance preoperative decision-making. METHODS: Pre-treatment magnetic resonance (MR) images of patients who underwent surgery between 2015 and 2019 with available WHO 2021 classification were used. The tumours were manually segmented on the T1w, T1-contrast enhanced, and T2w images using 3D Slicer. One hundred Pyradiomic features were extracted from each MR sequence. Models were built to classify (1) somatotroph and gonadotroph PitNETs and (2) high- and low-risk subtypes of non-functioning PitNETs. Feature were selected independently from the MR sequences and multi-sequence (combining data from more than one MR sequence) using Boruta and Pearson correlation. Support vector machine (SVM), logistic regression (LR), random forest (RF), and multi-layer perceptron (MLP) were the classifiers used. Data imbalance was addressed using the Synthetic Minority Oversampling TEchnique (SMOTE). Performance of the models were evaluated using area under the receiver operating curve (AUC), accuracy, sensitivity, and specificity. RESULTS: A total of 222 PitNET patients (train, n = 149; test, n = 73) were enrolled in this retrospective study. Multi-sequence-based LR model discriminated best between somatotroph and gonadotroph PitNETs, with a test AUC of 0.84, accuracy of 0.74, specificity of 0.81, and sensitivity of 0.70. Multi-sequence-based MLP model perfomed best for the high- and low-risk non-functioning PitNETs, achieving a test AUC of 0.76, accuracy of 0.67, specificity of 0.72, and sensitivity of 0.66. CONCLUSIONS: Utilizing pre-treatment MRI and radiomics holds promise for distinguishing high-risk from low-risk non-functioning PitNETs based on the latest WHO classification. This could assist neurosurgeons in making critical decisions regarding surgery or alternative management strategies for PitNETs after further clinical validation.

Multi-centre radiomics for prediction of recurrence following radical radiotherapy for head and neck cancers: Consequences of feature selection, machine learning classifiers and batch-effect harmonization.

Background and purpose: Radiomics models trained with limited single institution data are often not reproducible and generalisable. We developed radiomics models that predict loco-regional recurrence within two years of radiotherapy with private and public datasets and their combinations, to simulate small and multi-institutional studies and study the responsiveness of the models to feature selection, machine learning algorithms, centre-effect harmonization and increased dataset sizes. Materials and methods: 562 patients histologically confirmed and treated for locally advanced head-and-neck cancer (LA-HNC) from two public and two private datasets; one private dataset exclusively reserved for validation. Clinical contours of primary tumours were not recontoured and were used for Pyradiomics based feature extraction. ComBat harmonization was applied, and LASSO-Logistic Regression (LR) and Support Vector Machine (SVM) models were built. 95% confidence interval (CI) of 1000 bootstrapped area-under-the-Receiver-operating-curves (AUC) provided predictive performance. Responsiveness of the models' performance to the choice of feature selection methods, ComBat harmonization, machine learning classifier, single and pooled data was evaluated. Results: LASSO and SelectKBest selected 14 and 16 features, respectively; three were overlapping. Without ComBat, the LR and SVM models for three institutional data showed AUCs (CI) of 0.513 (0.481-0.559) and 0.632 (0.586-0.665), respectively. Performances following ComBat revealed AUCs of 0.559 (0.536-0.590) and 0.662 (0.606-0.690), respectively. Compared to single cohort AUCs (0.562-0.629), SVM models from pooled data performed significantly better at AUC = 0.680. Conclusions: Multi-institutional retrospective data accentuates the existing variabilities that affect radiomics. Carefully designed prospective, multi-institutional studies and data sharing are necessary for clinically relevant head-and-neck cancer prognostication models.

Framework for Machine Learning of CT and PET Radiomics to Predict Local Failure after Radiotherapy in Locally Advanced Head and Neck Cancers.

CONTEXT: Cancer Radiomics is an emerging field in medical imaging and refers to the process of converting routine radiological images that are typically qualitatively interpreted to quantifiable descriptions of the tumor phenotypes and when combined with statistical analytics can improve the accuracy of clinical outcome prediction models. However, to understand the radiomic features and their correlation to molecular changes in the tumor, first, there is a need for the development of robust image analysis methods, software tools and statistical prediction models which is often limited in low- and middle-income countries (LMIC). AIMS: The aim is to build a framework for machine learning of radiomic features of planning computed tomography (CT) and positron emission tomography (PET) using open source radiomics and data analytics platforms to make it widely accessible to clinical groups. The framework is tested in a small cohort to predict local disease failure following radiation treatment for head-and-neck cancer (HNC). The predictors were also compared with the existing Aerts HNC radiomics signature. SETTINGS AND DESIGN: Retrospective analysis of patients with locally advanced HNC between 2017 and 2018 and 31 patients with both pre- and post-radiation CT and evaluation PET were selected. SUBJECTS AND METHODS: Tumor volumes were delineated on baseline PET using the semi-automatic adaptive-threshold algorithm and propagated to CT; PyRadiomics features (total of 110 under shape/intensity/texture classes) were extracted. Two feature-selection methods were tested for model stability. Models were built based on least absolute shrinkage and selection operator-logistic and Ridge regression of the top pretreatment radiomic features and compared to Aerts' HNC-signature. Average model performance across all internal validation test folds was summarized by the area under the receiver operator curve (ROC). RESULTS: Both feature selection methods selected CT features MCC (GLCM), SumEntropy (GLCM) and Sphericity (Shape) that could predict the binary failure status in the cross-validated group and achieved an AUC >0.7. However, models using Aerts' signature features (Energy, Compactness, GLRLM-GrayLevelNonUniformity and GrayLevelNonUniformity-HLH wavelet) could not achieve a clear separation between outcomes (AUC = 0.51-0.54). CONCLUSIONS: Radiomics pipeline included open-source workflows which makes it adoptable in LMIC countries. Additional independent validation of data is crucial for the implementation of radiomic models for clinical risk stratification.

Accuracy of the freehand (fennell) technique using a uniform entry point and sagittal trajectory for insertion of thoracic pedicle screws: A computed tomography-based virtual simulation study.

BACKGROUND: Thoracic spine has complex pedicle anatomy with a narrow canal diameter which makes pedicle screw insertion challenging. Fennell et al. have described a simple freehand technique of thoracic pedicle screw placement. We have tested the accuracy of Fennell technique using computed tomography-based (CT-based) simulation model with pedicle screw simulator (PSS). METHODS: Normal CT thoracic spine obtained from CT thorax data of five patients were used in the 3D slicer environment using PSS for simulation. Entry points and axial trajectory as described by Fennell et al. and a sagittal trajectory parallel to the superior endplate were used for simulating the freehand technique using EA (entry angle) mode in the PSS. An ideal trajectory through the midsection of the pedicle from the same entry point and a sagittal trajectory parallel to the superior endplate were simulated using the ET (Entry Target) mode. Angle predicted by the software for an ideal axial trajectory was compared with the Fennell technique and this angle difference was noted at all the levels. Presence of pedicle breach was noted while simulating the Fennell technique. RESULTS: A total of 240 thoracic pedicle screw insertions were simulated, 120 screws by each technique. A sagittal trajectory parallel to the superior endplate caused no pedicle breach in the cranial-caudal direction at any level. No medial or lateral breach was noted while using an axial trajectory of 30° at T1-T2 and 20° from T3-T10. A 20° axial trajectory at T11 and T12 resulted in a breach of the medial cortex and the ideal mean axial angles at T11 and T12 were 2.8° and 6.5°, respectively. CONCLUSIONS: Fennell technique was effectively simulated using PSS. A uniform entry point and sagittal trajectory parallel to the superior endplate serves as a useful guide for freehand insertion of thoracic pedicle screws. At T11 and 12, ideal axial trajectories are less than 10°.

Hybrid positron emission tomography segmentation of heterogeneous lung tumors using 3D Slicer: improved GrowCut algorithm with threshold initialization.

This paper presents an improved GrowCut (IGC), a positron emission tomography-based segmentation algorithm, and tests its clinical applicability. Contrary to the traditional method that requires the user to provide the initial seeds, the IGC algorithm starts with a threshold-based estimate of the tumor and a three-dimensional morphologically grown shell around the tumor as the foreground and background seeds, respectively. The repeatability of IGC from the same observer at multiple time points was compared with the traditional GrowCut algorithm. The algorithm was tested in 11 nonsmall cell lung cancer lesions and validated against the clinician-defined manual contour and compared against the clinically used 25% of the maximum standardized uptake value [SUV-(max)], 40% [Formula: see text], and adaptive threshold methods. The time to edit IGC-defined functional volume to arrive at the gross tumor volume (GTV) was compared with that of manual contouring. The repeatability of the IGC algorithm was very high compared with the traditional GrowCut ([Formula: see text]) and demonstrated higher agreement with the manual contour with respect to threshold-based methods. Compared with manual contouring, editing the IGC achieved the GTV in significantly less time ([Formula: see text]). The IGC algorithm offers a highly repeatable functional volume and serves as an effective initial guess that can well minimize the time spent on labor-intensive manual contouring.

Targeting apoptosis for optical imaging of infection.

PURPOSE: Infection is ubiquitous and a major cause of morbidity and mortality. The most reliable method for localizing infection requires radiolabeling autologous white blood cells ex vivo. A compound that can be injected directly into a patient and can selectively image infectious foci will eliminate the drawbacks. The resolution of infection is associated with neutrophil apoptosis and necrosis presenting phosphatidylserine (PS) on the neutrophil outer leaflet. Targeting PS with intravenous administration of a PS-specific, near-infrared (NIR) fluorophore will permit localization of infectious foci by optical imaging. METHODS: Bacterial infection and sterile inflammation were induced in separate groups (n = 5) of mice. PS was targeted with a NIR fluorophore, PSVue(®)794 (2.7 pmol). Imaging was performed (ex = 730 nm, em = 830 nm) using Kodak Multispectral FX-Pro system. The contralateral normal thigh served as an individualized control. Confocal microscopy of normal and apoptotic neutrophils and bacteria confirmed PS specificity. RESULTS: Lesions, with a 10-s image acquisition, were unequivocally visible at 5 min post-injection. At 3 h post-injection, the lesion to background intensity ratios in the foci of infection (6.6 ± 0.2) were greater than those in inflammation (3.2 ± 0.5). Image fusions confirmed anatomical locations of the lesions. Confocal microscopy determined the fluorophore specificity for PS. CONCLUSIONS: Targeting PS presented on the outer leaflet of apoptotic or necrotic neutrophils as well as gram-positive microorganism with PS-specific NIR fluorophore provides a sensitive means of imaging infection. Literature indicates that NIR fluorophores can be detected 7-14 cm deep in tissue. This observation together with the excellent results and the continued development of versatile imaging devices could make optical imaging a simple, specific, and rapid modality for imaging infection.

Apoptotic abscess imaging with 99mTc-HYNIC-rh-Annexin-V.

Abscess formation causes systemic and localized up-regulation of neutrophil [polymorphonuclear leukocytes (PMNs)] signaling pathways. In the abscess, following bacterial ingestion or PMN activation by inflammatory mediators, PMN apoptosis is elevated and leads to the externalization of phosphatidylserine. Annexin-V (AnxV) has been shown to have high affinity to externalized phosphatidylserine. We hypothesized that (99m)Tc-AnxV will target high densities of apoptotic PMNs and image abscesses. AnxV, conjugated with hydrazinenicaotinamide (HYNIC), was labeled with reduced (99m)TcO(4)(-) and its purity was determined by instant thin-layer chromatography. Apoptosis was induced in isolated human PMNs by incubation in 2% saline for 17 and 22 h at 37 degrees C. PMNs were then incubated with (99m)Tc-HYNIC-AnxV and associated (99m)Tc was determined. Abscesses were induced in mice by intramuscular injection of bacteria or turpentine. Following intravenous administration of (99m)Tc-HYNIC-AnxV, mice were imaged and tissue distribution studied at 4 and 24 h. Radiochemical purity of (99m)Tc-HYNIC-AnxV was 84.9+/-8.11%. At 17 h, (99m)Tc-HYNIC-AnxV bound to apoptotic PMNs was 71.6+/-0.01% and 48.6+/-0.01% for experimental and control cells, respectively (P=.002). At 22 h, experimental cells retained 74.9+/-0.02% and control cells retained 47.2+/-0.02% (P=.005). (99m)Tc-HYNIC-AnxV associated with bacterial abscesses was 1.25+/-0.09 and 3.75+/-0.83 percent injected dose per gram (%ID/g) at 4 and 24 h compared to turpentine abscesses which was 1.02+/-0.16 and 0.72+/-0.17 %ID/g at 4 (P