Ensemble detection of hand joint ankylosis and subluxation in radiographic images using deep neural networks.

The modified total Sharp score (mTSS) is often used as an evaluation index for joint destruction caused by rheumatoid arthritis. In this study, special findings (ankylosis, subluxation, and dislocation) are detected to estimate the efficacy of mTSS by using deep neural networks (DNNs). The proposed method detects and classifies finger joint regions using an ensemble mechanism. This integrates multiple DNN detection models, specifically single shot multibox detectors, using different training data for each special finding. For the learning phase, we prepared a total of 260 hand X-ray images, in which proximal interphalangeal (PIP) and metacarpophalangeal (MP) joints were annotated with mTSS by skilled rheumatologists and radiologists. We evaluated our model using five-fold cross-validation. The proposed model produced a higher detection accuracy, recall, precision, specificity, F-value, and intersection over union than individual detection models for both ankylosis and subluxation detection, with a detection rate above 99.8% for the MP and PIP joint regions. Our future research will aim at the development of an automatic diagnosis system that uses the proposed mTSS model to estimate the erosion and joint space narrowing score.

Predicting human chronological age via AI analysis of dorsal hand versus facial images: A study in a cohort of Indian females.

Predicting a person's chronological age (CA) from visible skin features using artificial intelligence (AI) is now commonplace. Often, convolutional neural network (CNN) models are built using images of the face as biometric data. However, hands hold telltale signs of a person's age. To determine the utility of using only hand images in predicting CA, we developed two deep CNNs based on 1) dorsal hand images (H) and 2) frontal face images (F). Subjects (n = 1454) were Indian women, 20-80 years, across three geographic cohorts (Mumbai, New Delhi and Bangalore) and having a broad variation in skin tones. Images were randomised: 70% of F and 70% of H were used to train CNNs. The remaining 30% of F and H were retained for validation. CNN validation showed mean absolute error for predicting CA using F and H of 4.1 and 4.7 years, respectively. In both cases correlations of predicted and actual age were statistically significant (r(F) = 0.93, r(H) = 0.90). The CNNs for F and H were validated for dark and light skin tones. Finally, by blurring or accentuating visible features on specific regions of the hand and face, we identified those features that contributed to the CNN models. For the face, areas of the inner eye corner and around the mouth were most important for age prediction. For the hands, knuckle texture was a key driver for age prediction. Collectively, for AI estimates of CA, CNNs based solely on hand images are a viable alternative and comparable to CNNs based on facial images.

Relationship Between Hand Function and Handheld Ultrasound Imaging in Inclusion Body Myositis.

OBJECTIVE: Ultrasound studies in inclusion body myositis (IBM) have reported a characteristic pattern of increased echointensity in the flexor digitorum profundus (FDP) with relative sparing of the flexor carpi ulnaris (FCU). We examined the relationship between echointensity of the FDP and FCU muscles and hand strength or patient-reported outcomes (PROs). METHODS: A total of 15 patients with IBM were recruited. Ultrasound images of the FDP and FCU muscles were obtained by a point-of-care ultrasound and graded using the modified Heckmatt score. Hand grip and neutral pinch strength were measured by dynamometry. PROs were assessed by the IBM Upper Extremity Function Scale. RESULTS: FDP and/or FCU modified Heckmatt score showed a significant relationship with grip, neutral pinch strength, and PROs. CONCLUSIONS: Point-of-care ultrasound examination of the forearm may serve as an extension of the neuromuscular examination. The semi-qualitative echointensity rating based on modified Heckmatt score seems to correlate well with the objective strength measurement and PROs.

Lipofibromatous Hamartoma of the Median Nerve - A Rare Condition in the Hand.

Lipofibromatous hamartoma (LFH) of the median nerve is a rare condition in the hand and often remains asymptomatic for a significant period. MRI imaging can reveal unique tumour characteristics; however, the definitive diagnosis is confirmed through a tissue biopsy. In this report, a 38-year-old male presented with a gradually growing mass on his right hand. Physical examination revealed a large soft tissue mass extending from the thenar area to the wrist, causing compression of the median nerve. MRI confirmed the presence of a distinct soft tissue mass on the volar side of the hand. The mass was excised along with a fascicle and confirmed by histological examination. One year after surgery, sensation has improved, but weakness remains and opponensplasty was offered to the patient. Although the treatment strategy of LFH of the median nerve remains controversial, delayed treatment can result in severe compressive neuropathy and irreversible nerve damage. Level of Evidence: Level V (Therapeutic).

The Addition of Hand-specific Skeletal Maturity Parameters Does Not Improve Skeletal Maturity Estimation Accuracy of the Modified Fels Wrist System.

BACKGROUND: The Modified Fels Wrist system is potentially the most accurate clinically accessible skeletal maturity system utilizing hand or wrist radiographs. During development, parameters distal to the metacarpals were excluded. We attempted to further optimize the Modified Fels wrist system through the inclusion of hand parameters distal to the metacarpals. METHODS: Forty-three new anteroposterior (AP) hand radiographic parameters were identified from the Fels and Greulich and Pyle (GP) skeletal maturity systems. Twelve parameters were eliminated from further evaluation for poor correlation with skeletal maturity, poor reliability, and lack of relevance in the peripubertal years. In addition to the 8 previously described Modified Fels Wrist parameters, 31 hand radiographic parameters were evaluated on serial peripubertal AP hand radiographs to identify the ones most important for accurately estimating skeletal age. This process produced a "Modified Fels hand-wrist" model; its performance was compared with (1) GP only; (2) Sanders Hand (SH) only; (3) age, sex, and GP; (4) age, sex, and SH; and (5) Modified Fels Wrist system. RESULTS: Three hundred seventy-two radiographs from 42 girls and 38 boys were included. Of the 39 radiographic parameters that underwent full evaluation, 9 remained in the combined Modified Fels Hand-Wrist system in addition to chronological age and sex. Four parameters are wrist specific, and the remaining 5 are hand specific. The Hand-Wrist system outperformed both GP and SH in estimating skeletal maturity ( P <0.001). When compared with the Modified Fels Wrist system, the Modified Fels Hand-Wrist system performed similarly regarding skeletal maturity estimation (0.36±0.32 vs. 0.34±0.26, P =0.59) but had an increased (worse) rate of outlier predictions >1 year discrepant from true skeletal maturity (4.9% vs. 1.9%, P =0.01). CONCLUSIONS: The addition of hand parameters to the existing Modified Fels Wrist system did not improve skeletal maturity estimation accuracy and worsened the rate of outlier estimations. When an AP hand-wrist radiograph is available, the existing Modified Fels wrist system is best for skeletal maturity estimation. LEVEL OF EVIDENCE: Level III.

The Implementation of a Gesture Recognition System with a Millimeter Wave and Thermal Imager.

During the COVID-19 pandemic, the number of cases continued to rise. As a result, there was a growing demand for alternative control methods to traditional buttons or touch screens. However, most current gesture recognition technologies rely on machine vision methods. However, this method can lead to suboptimal recognition results, especially in situations where the camera is operating in low-light conditions or encounters complex backgrounds. This study introduces an innovative gesture recognition system for large movements that uses a combination of millimeter wave radar and a thermal imager, where the multi-color conversion algorithm is used to improve palm recognition on the thermal imager together with deep learning approaches to improve its accuracy. While the user performs gestures, the mmWave radar captures point cloud information, which is then analyzed through neural network model inference. It also integrates thermal imaging and palm recognition to effectively track and monitor hand movements on the screen. The results suggest that this combined method significantly improves accuracy, reaching a rate of over 80%.

Diagnosis and treatment of flexor tendon injuries of the hand: what the radiologist needs to know.

This article reviews the diagnosis and treatment of flexor tendon injuries of the hand highlighting flexor tendon anatomy, important pre-operative imaging findings, surgical options, and post-operative complications. Imaging plays a key role in guiding treatment of these difficult to manage injuries. Thus, it is important for radiologists to have a sound understanding of factors important in treatment decision-making. In the pre-operative setting, accurately identifying the location of the torn proximal tendon stump in subacute and chronic injuries helps dictate whether the patient is a candidate for a primary flexor tendon repair or may require a tendon reconstruction to restore function. In the post-operative setting, the status of the repair and presence of surrounding adhesions help dictate if and when the patient will require subsequent surgery and whether that surgery will be a tenolysis, revision repair, reconstruction, or fusion.

High Performance Wearable Ultrasound as a Human-Machine Interface for Wrist and Hand Kinematic Tracking.

OBJECTIVE: Non-invasive human machine interfaces (HMIs) have high potential in medical, entertainment, and industrial applications. Traditionally, surface electromyography (sEMG) has been used to track muscular activity and infer motor intention. Ultrasound (US) has received increasing attention as an alternative to sEMG-based HMIs. Here, we developed a portable US armband system with 24 channels and a multiple receiver approach, and compared it with existing sEMG- and US-based HMIs on movement intention decoding. METHODS: US and motion capture data was recorded while participants performed wrist and hand movements of four degrees of freedom (DoFs) and their combinations. A linear regression model was used to offline predict hand kinematics from the US (or sEMG, for comparison) features. The method was further validated in real-time for a 3-DoF target reaching task. RESULTS: In the offline analysis, the wearable US system achieved an average [Formula: see text] of 0.94 in the prediction of four DoFs of the wrist and hand while sEMG reached a performance of [Formula: see text]= 0.60. In online control, the participants achieved an average 93% completion rate of the targets. CONCLUSION: When tailored for HMIs, the proposed US A-mode system and processing pipeline can successfully regress hand kinematics both in offline and online settings with performances comparable or superior to previously published interfaces. SIGNIFICANCE: Wearable US technology may provide a new generation of HMIs that use muscular deformation to estimate limb movements. The wearable US system allowed for robust proportional and simultaneous control over multiple DoFs in both offline and online settings.

Two arterial variations of the hand and wrist present bilaterally (persistent median artery and superficial dorsal branch of the radial artery): ultrasound findings.

PURPOSE: Arterial variations of the upper limb may bear high importance for many clinical procedures, including the use of flaps in plastic surgery. We present a feasible way for visualization and confirmation of presence of these variations. METHODS: All variations were detected by ultrasonography and confirmed by Color Doppler Imaging. Proper documentation was taken in order to present our findings. RESULTS: We report a case of a 19-year-old female who showed two concomitant arterial variations of the forearm and the hand bilaterally. These two variations were the persistent median artery and the superficial dorsal branch of the radial artery which both significantly contributed to the blood supply of the hand. All examinations were performed by the same investigator and all findings were reviewed by an experienced sonographist. CONCLUSION: An unusual arrangement of the arterial system can be easily detected. We present a feasible way to prevent iatrogenic injuries and increase utilization of anatomical variants knowledge in surgery by using ultrasound prior to planning surgical procedures.

The lateral approach water bath: A novel method of ultrasound imaging of the hand.

BACKGROUND: Traditional water baths for ultrasound exams place a hand into a pan of water and submerge an ultrasound probe into the water. While this improves ultrasound transmission and moves structures into the focal zone to make higher resolution images, this method does have limitations. Patients must be manipulated directly under the probe, which can be limited by pain or normal movement restrictions. The probe must also be held very still in water to minimize motion artifact. The lateral approach water bath method addresses such limitations by imaging through the side of a thin-walled plastic container without submerging the probe. This reduces much need for patient manipulation by imaging through the side of a column-shaped bath, which has 360 degrees of imaging freedom. It also stabilizes the probe directly against the flat, firm container to reduce image degrading motion artifact. We hypothesized that because of these improvements the lateral approach water bath might create higher quality images than traditional water baths. METHODS: We compared twenty images from each method, which were obtained with the same model and ultrasound operator at the same time. Two ultrasound fellowship trained blinded reviewers rated the images for quality and adequacy for clinical decision making on a scale from 1 to 5. RESULTS: Image quality was better for the lateral water bath, with an average rating of 4.2 compared to the traditional bath's 2.6 (p < 0.001). Adequacy to aid clinical decision making was better for the lateral approach bath with an average rating of 4.0 compared to the traditional bath's 2.6 (p < 0.001). The lateral bath also had a smaller range for image quality and thus greater consistency. CONCLUSIONS: The lateral approach water bath is a method of hand imaging that produces higher quality, more consistent, and more clinically useful images than traditional water bath imaging.

Whole-hand and regional bone mineral density involvement in rheumatoid arthritis.

Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by symmetric polyarthritis that can lead to joint deformity, disability, and osteoporosis. We aimed to evaluate whole hand and regional BMD in RA patients compared to controls. In addition, we evaluated the BMD of dominant versus non-dominant hands in healthy subjects. We included adult female and male RA patients and control subjects matched by age, sex, and BMI. BMD (g/cm2) was measured by DXA in lumbar spine (LS), whole hand, and three regions of interest: carpus, metacarpal bones, and phalanges. Results: 44 control subjects (49.5±11.8 y) and 60 with RA (52.7±12.7 y) were included. Significant lower BMD in RA patients was found in LS (-8.7%), dominant whole hand (-9.5%), carpus, metacarpal bones, and phalanges, and non-dominant whole hand (-8.7%), metacarpal bones, and phalanges compared to controls. A significant positive correlation was found between LS and whole-hand BMD (dominant r=.63, non-dominant r=.67). Finally, the whole hand, metacarpal bones, and carpus BMD measurements were significantly higher in the dominant hand compared to the non-dominant hand without differences in the phalangeal ROI. In conclusion, hand BMD was significantly lower in RA patients compared to control subjects and there was a significant correlation with LS BMD. We demonstrated that BMD measurements of the whole-hand, and different ROI (carpus, metacarpal bones, and phalanges) by DXA would be an easily reproducible technique to evaluate bone loss. In addition, the whole hand, metacarpal bones and carpus BMD measurements were significantly higher in the dominant hand compared to the non-dominant hand without differences in the phalanges.

Whole-hand and regional bone mineral density involvement in rheumatoid arthritis / Compromiso de la densidad mineral ósea de la mano completa y regional en la artritis reumatoide

Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by symmetric polyarthritis that can lead to joint deformity, disability, and osteoporosis. We aimed to evaluate whole hand and regional BMD in RA patients compared to controls. In addition, we evaluated the BMD of dominant versus non-dominant hands in healthy subjects. We included adult female and male RA patients and control subjects matched by age, sex, and BMI. BMD (g/cm2) was measured by DXA in lumbar spine (LS), whole hand, and three regions of interest: carpus, metacarpal bones, and phalanges. Results: 44 control subjects (49.5±11.8 y) and 60 with RA (52.7±12.7 y) were included. Significant lower BMD in RA patients was found in LS (−8.7%), dominant whole hand (−9.5%), carpus, metacarpal bones, and phalanges, and non-dominant whole hand (−8.7%), metacarpal bones, and phalanges compared to controls. A significant positive correlation was found between LS and whole-hand BMD (dominant r=.63, non-dominant r=.67). Finally, the whole hand, metacarpal bones, and carpus BMD measurements were significantly higher in the dominant hand compared to the non-dominant hand without differences in the phalangeal ROI. In conclusion, hand BMD was significantly lower in RA patients compared to control subjects and there was a significant correlation with LS BMD. We demonstrated that BMD measurements of the whole-hand, and different ROI (carpus, metacarpal bones, and phalanges) by DXA would be an easily reproducible technique to evaluate bone loss. In addition, the whole hand, metacarpal bones and carpus BMD measurements were significantly higher in the dominant hand compared to the non-dominant hand without differences in the phalanges.(AU)

La artritis reumatoide (AR) es una enfermedad autoinmune crónica caracterizada por poliartritis simétrica que puede provocar deformidad e incapacidad articular y osteoporosis. Nuestro objetivo fue evaluar la DMO de manos completa y por regiones en los pacientes con AR en comparación con los controles. Se incluyeron pacientes adultos de ambos sexos con AR, y sujetos controles de edad, sexo e IMC similar. La DMO se midió por DXA en columna lumbar (CL), manos completas y 3 regiones de interés: carpo, metacarpianos y falanges. Resultados: se incluyeron 44 sujetos control (49,5±11,8 años) y 60 con AR (52,7±12,7 años). Se encontró una DMO significativamente más baja en los pacientes con AR en CL (−8,7%), mano completa dominante (−9,5%) y mano completa no dominante (−8,7%) en comparación con los sujetos controles. Se encontró una correlación positiva significativa entre la CL y la DMO de la mano completa (dominante, r=0,63; no dominante, r=0,67). Finalmente, la DMO de la mano completa, los huesos metacarpianos y el carpo fueron significativamente más altos en la mano dominante en comparación con la mano no dominante sin diferencias en la región de las falanges. En conclusión, la DMO de la mano fue significativamente menor en los pacientes con AR en comparación con los sujetos controles, y hubo una correlación significativa con la DMO de la CL. Demostramos que las mediciones de la DMO de toda la mano y diferentes ROI (carpo, huesos metacarpianos y falanges) por DXA serían una técnica fácilmente reproducible para evaluar la pérdida ósea. Además, la DMO de la mano completa, los huesos metacarpianos y el carpo fueron significativamente más altos en la mano dominante en comparación con la mano no dominante.(AU)

An Improved Multimodal Biometric Identification System Employing Score-Level Fuzzification of Finger Texture and Finger Vein Biometrics.

This research work focuses on a Near-Infra-Red (NIR) finger-images-based multimodal biometric system based on Finger Texture and Finger Vein biometrics. The individual results of the biometric characteristics are fused using a fuzzy system, and the final identification result is achieved. Experiments are performed for three different databases, i.e., the Near-Infra-Red Hand Images (NIRHI), Hong Kong Polytechnic University (HKPU) and University of Twente Finger Vein Pattern (UTFVP) databases. First, the Finger Texture biometric employs an efficient texture feature extracting algorithm, i.e., Linear Binary Pattern. Then, the classification is performed using Support Vector Machine, a proven machine learning classification algorithm. Second, the transfer learning of pre-trained convolutional neural networks (CNNs) is performed for the Finger Vein biometric, employing two approaches. The three selected CNNs are AlexNet, VGG16 and VGG19. In Approach 1, before feeding the images for the training of the CNN, the necessary preprocessing of NIR images is performed. In Approach 2, before the pre-processing step, image intensity optimization is also employed to regularize the image intensity. NIRHI outperforms HKPU and UTFVP for both of the modalities of focus, in a unimodal setup as well as in a multimodal one. The proposed multimodal biometric system demonstrates a better overall identification accuracy of 99.62% in comparison with 99.51% and 99.50% reported in the recent state-of-the-art systems.

Occupational radiation dose on the hand of assisting medical staff in diagnostic CT scans.

Chronic radiation exposure increases the risk of skin damage of medical personnel engaged in radiology. However, hand dose measurements in computed tomography (CT) for diagnostic purposes have not been evaluated. The occupational radiation dose to the hands of CT assistants was herein investigated to evaluate its compliance with the equivalent dose limit for the hand (500 mSv/year). The occupational doses of nine CT assistants were measured in 89 cases (April 2017-May 2018) by installing radio-photoluminescence glass dosemeters (GD-302 M) (70-µm dose-equivalent conversion coefficient = 0.37) on the dorsal aspect of both hands. The occupational dose to the hand was the highest with head holding (right: 1.14 mSv/CT scan, left: 1.07 mSv/CT scan). Considering the results for annual work, even for head holding, the hand dose of the CT-assisting personnel was insignificant. However, CT assistants should be mindful of the possibility of locally higher doses to hands.

Measurement of the lumbrical muscle activity of the hand using electromyography supported by the ultrasound imaging technique with string navigation.

Although placing surface electrodes on small muscles by palpation is difficult, ultrasound guidance may enable electrode placement on the small muscles. This study aimed to examine whether ultrasound guidance is helpful for placement of electrodes on a small muscle, such as the hand lumbrical muscle. Twelve dominant hands of 12 healthy right-handed adults were included in this study. The first lumbrical muscle belly of the hands was identified using ultrasound guidance with a string navigation technique for placing surface electrodes. This technique was designed to identify the location of the center of the muscle belly under ultrasound imaging using a string. After the electrodes were placed on the muscle belly using this technique, the surface electromyographic signals of the first lumbrical, first dorsal interosseous, and adductor pollicis muscles were recorded. The activity of the lumbrical muscle could be separately measured of the first dorsal interosseous and adductor pollicis muscles. This technique has the potential to enable surface electromyography of small muscles for which placement of surface electrodes by palpation is challenging.

MuTr: Multi-Stage Transformer for Hand Pose Estimation from Full-Scene Depth Image.

This work presents a novel transformer-based method for hand pose estimation-DePOTR. We test the DePOTR method on four benchmark datasets, where DePOTR outperforms other transformer-based methods while achieving results on par with other state-of-the-art methods. To further demonstrate the strength of DePOTR, we propose a novel multi-stage approach from full-scene depth image-MuTr. MuTr removes the necessity of having two different models in the hand pose estimation pipeline-one for hand localization and one for pose estimation-while maintaining promising results. To the best of our knowledge, this is the first successful attempt to use the same model architecture in standard and simultaneously in full-scene image setup while achieving competitive results in both of them. On the NYU dataset, DePOTR and MuTr reach precision equal to 7.85 mm and 8.71 mm, respectively.

Fast and Accurate ROI Extraction for Non-Contact Dorsal Hand Vein Detection in Complex Backgrounds Based on Improved U-Net.

In response to the difficulty of traditional image processing methods to quickly and accurately extract regions of interest from non-contact dorsal hand vein images in complex backgrounds, this study proposes a model based on an improved U-Net for dorsal hand keypoint detection. The residual module was added to the downsampling path of the U-Net network to solve the model degradation problem and improve the feature information extraction ability of the network; the Jensen-Shannon (JS) divergence loss function was used to supervise the final feature map distribution so that the output feature map tended to Gaussian distribution and improved the feature map multi-peak problem; and Soft-argmax is used to calculate the keypoint coordinates of the final feature map to realize end-to-end training. The experimental results showed that the accuracy of the improved U-Net network model reached 98.6%, which was 1% better than the original U-Net network model; the improved U-Net network model file was only 1.16 M, which achieved a higher accuracy than the original U-Net network model with significantly reduced model parameters. Therefore, the improved U-Net model in this study can realize dorsal hand keypoint detection (for region of interest extraction) for non-contact dorsal hand vein images and is suitable for practical deployment in low-resource platforms such as edge-embedded systems.

Prediction of Fishman's skeletal maturity indicators using artificial intelligence.

The present study aimed to evaluate the performance of automated skeletal maturation assessment system for Fishman's skeletal maturity indicators (SMI) for the use in dental fields. Skeletal maturity is particularly important in orthodontics for the determination of treatment timing and method. SMI is widely used for this purpose, as it is less time-consuming and practical in clinical use compared to other methods. Thus, the existing automated skeletal age assessment system based on Greulich and Pyle and Tanner-Whitehouse3 methods was further developed to include SMI using artificial intelligence. This hybrid SMI-modified system consists of three major steps: (1) automated detection of region of interest; (2) automated evaluation of skeletal maturity of each region; and (3) SMI stage mapping. The primary validation was carried out using a dataset of 2593 hand-wrist radiographs, and the SMI mapping algorithm was adjusted accordingly. The performance of the final system was evaluated on a test dataset of 711 hand-wrist radiographs from a different institution. The system achieved a prediction accuracy of 0.772 and mean absolute error and root mean square error of 0.27 and 0.604, respectively, indicating a clinically reliable performance. Thus, it can be used to improve clinical efficiency and reproducibility of SMI prediction.

Imaging of Anatomical Variants Around the Wrist and Hand.

Wrist and hand anatomy may present several clinically relevant variants that involve bones, muscles, tendons, and nerves. Thorough knowledge of these abnormalities and their appearance in imaging studies is useful for proper management. In particular, it is necessary to differentiate the incidental findings that do not represent a trigger for a specific syndrome from those anomalies causing symptoms and functional impairment. This review reports the most common anatomical variants encountered in clinical practice and briefly discusses their embryogenesis, related clinical syndrome if present, and their appearance using different imaging techniques. The information each diagnostic study (ultrasonography, radiographs, computed tomography, and magnetic resonance imaging) may provide is described for each condition.