Objective:To evaluate the application value of artificial intelligence (AI) and robot of stereotactic surgery (ROSS)-assisted ultra-early thalamic hematoma drainage in elderly patients with minor thalamus and internal capsule hematoma (TICH) and severe hypoperfusion in the internal capsule area.Methods:A retrospective cross-sectional study was performed; 278 patients with TICH and severe hypoperfusion in the internal capsule area identified by AI medical imaging diagnostic system were enrolled from Department of Neurosurgery, Xingtai Central Hospital from January 2023 to August 2024. Among them, 134 patients (study group) received AI and ROSS-assisted ultra-early thalamic hematoma drainage, and 144 patients (control group) received drug treatment. Differences in baseline data and therapy efficiency between the two groups of patients were compared. Multivariate Logistic regression analysis was used to screen the independent influencing factors for prognosis in the study group 180 days after onset (good prognosis: modified Rankin scale scores of 0-3).Results:Compared with the control group, the study group had significantly higher Glasgow coma scale (GCS) score and regional cerebral blood flow (rCBF) in the internal capsule area 14 days after onset, and statistically higher therapy efficiency and good prognosis rate 180 days after onset ( P<0.05). In the study group, 84 patients had a good prognosis and 50 had a poor prognosis 180 days after onset; compared with the good prognosis group, the poor prognosis group had significantly higher proportion of patients with hyper-homocysteinemia history, lower GCS score on admission, larger volume of thalamic hematoma, higher proportions of patients with combined intraventricular hemorrhage and shunt-dependent hydrocephalus, and lower rCBF in the internal capsule area on admission ( P<0.05). Multivariate Logistic regression analysis showed that thalamic hematoma volume ( OR=2.527, 95% CI: 1.504-4.247, P<0.001), combined intraventricular hemorrhage ( OR=2.325, 95% CI: 1.460-3.703, P<0.001), shunt dependent hydrocephalus ( OR=2.371, 95% CI: 1.267-4.078, P=0.006), and rCBF in the internal capsule area on admission ( OR=0.672, 95% CI: 0.314-1.025, P<0.001) were independent risk factors for prognosis of patients in the study group. Conclusion:AI and ROSS-assisted ultra-early thalamic hematoma drainage is effective for elderly patients with minor TICH and severe hypoperfusion in the internal capsule area; patients with large thalamic hematoma volume, combined intraventricular hemorrhage, shunt-dependent hydrocephalus, and low rCBF in the internal capsule area on admission are prone to have a poor prognosis.

Objective To investigate the effectiveness and safety of infarct core resection combined with decompressive craniectomy(DC)based on corticospinal tract(CST)protection in the treatment of massive cerebral infarction(MCI)with malignant brain edema.Methods This study retrospectively enrolled MCI patients with malignant brain edema who underwent internal decompression combined with DC at Xingtai Central Hospital from January 2021 to June 2024.The enrolled patients were divided into a control group and an experimental group base on the intracranial internal decompression method used.All patients underwent CT perfusion(CTP),CT angiography(CTA),diffusion-weighted imaging(DWI),and diffusion tensor imaging(DTI)within 24 h of admission.Preoperative imaging data was automatically processed using an artificial intelligence diagnostic system.For the experimental group,the imaging data was fused within a neuro-navigation system preoperatively to visualize the spatial relationships between the infarct core,ischemic penumbra,and CST and infarct core resection combined with DC was performed while protecting the CST through neuro-navigation.The control group underwent anterior temporal lobectomy combined with DC.Baseline and clinical data were collected from both groups,including gender,age,smoking history,alcohol consumption history,diabetes,hypertension,hyperlipidemia,hyperhomocysteinemia,atrial fibrillation history,responsible occluded vessel(internal carotid artery,middle cerebral artery),preoperative infarct volume on DWI,preoperative ischemic penumbra volume,preoperative the National Institutes of Health stroke scale(NIHSS)score,time from onset to surgery,intraoperative procedure duration,intraoperative blood loss,preoperative and 1-month postoperative fraction anisotropy(FA)values of the CST on the affected side,modified Rankin scale(mRS)score at 6 months postoperatively,and surgery-related complications within 1 month postoperatively(intracranial hemorrhage[operative site oozing,hemorrhagic transformation]and intracranial infection[surgical incision site infection,empyema,brain abscess,meningitis]).6-month follow-up after surgery were conducted through outpatient visit or telephone calls and prognosis of patients was evaluated using the mRS(with mRS of 0-3 defined as good prognosis,4-6 as poor prognosis,and 6 indicating death).The effectiveness indicators included FA value of the affected CST at 1 month postoperatively,good prognosis rate after surgery at 6 months,and 6-month mortality rate after surgery.The safety indicators included the incidence rates of surgical complications(intracranial hemorrhage and infection)within 1 month postoperatively.Based on preoperative DTI images,all patients were further divided into a CST-intact(infarct core did not invade CST,CST morphology intact or deformed/shifted)and a CST-damaged(infarct core invaded CST,CST disrupted or interrupted)subgroup for analysis.Results A total of 62patients(37 males,25 females,age 49-60 years,mean[55±4]years)were enrolled in this study.With 28 patients in the experimental group and 34 in the control group.(1)No significant differences were found in baseline or clinical data between the experimental and control groups(all P＞0.05),and the reoperative FA values of the affected CST were showed no significant differences(P=0.588).(2)The efficacy and safety metrics were evaluated.For the efficacy indices,at 1 month after the surgery,FA values of the affected CST increased significantly compared to preoperative values in both groups(0.409±0.051 vs.0.312±0.052 in the experimental group,and,0.381±0.048 vs.0.319±0.049 in control group;both P＜0.05),and the FA value was significantly higher in the experimental group than that in the control group(0.409±0.051 vs.0.381±0.048,P=0.030).At the 6-month follow-ups,the good prognosis rate was significantly higher in the experimental group than that in the control group(39.3％[11/28]vs.14.7％[5/34],P=0.028).No significant difference in the 6-month mortality rate were observed between the two groups(P=0.787).For the safety indices,no significant differences were found in the incidence rates of intracranial hemorrhage or intracranial infection within 1 month postoperatively between the two groups(both P＞0.05).(3)For further subgroup analysis,no significant differences were found in baseline or clinical data between the CST-damaged subgroup and the CST-intact subgroup in both the experimental and control groups(all P＞0.05).In CST-intact subgroup,FA values of the affected CST increased significantly at 1 month postoperatively compared to preoperatively in the study group(0.428±0.047 vs.0.342±0.045,P＜0.05)and the control group(0.401±0.051 vs.0.347±0.048,P＜0.05).While in the CST-damaged subgroup,no significant differences were found in FA value of the affected CST 1 month postoperatively compared with that preoperatively in both the experimental and control groups(bothP＞0.05).A significantly higher FA values 1 month postoperatively(0.428±0.047 vs.0.401±0.051,P=0.036)and good prognosis rate(9/12 vs.4/16,P=0.020)were observed in the CST-intact subgroup of the experimental group comparing with that of the control group,while there was no statistically significant difference in the 6-month mortality rate between the groups within the CST-intact subgroup(P=1.000).There were no statistically significant differences between the experimental group and the control group in both efficacy and safety indices within the CST-damaged subgroup(all P＞0.05).Conclusions Infarct core resection combining DC with CST protection demonstrates superior neurological functional improvement in comparison with anterior temporal lobectomy combining DC in treating MCI with malignant brain edema,particularly for patients with an intact CST before surgery(as indicated in patients'preoperative imaging results).This(infarct core resection combining DC with CST protection)approach does not increase the incidence of surgical complications.Prospective large sample controlled studies are required for further validation.

Objective To investigate the clinical efficacy of internal decompression based on white matter tract preservation in the treatment of malignant middle cerebral artery infarction(MMCAI).Methods A retrospective analysis was conducted on 54 patients with MMCAI.Patients were divided into a study group(n=26)and a control group(n=28)according to the surgical approach.Patients in the study group underwent preoperative fusion of CT,CTP,DWI,and DTI imaging data within a neuronavigation system.This fusion visualized the spatial relationships between the infarct core(IC),ischemic penumbra,and the corticospinal tract(CST).Subsequently,IC resection combined with decompressive craniectomy(DC)was performed while protecting the CST.Patients in the control group underwent DC alone.Key outcome measures included:changes in fractional anisotropy(FA)within the affected CST projection area at 1 month postoperatively;and 6-month postoperative mRS score,mortality,and surgical complications at 6 months postoperatively.Results At 1 month postoperatively,FA in the affected CST projection area were significantly higher in the study group than in the control group(0.092±0.013 vs.0.082±0.008,P<0.05).At the 6-month follow-up,the postoperative mRS score in the study group was significantly lower than that in the control group[2.3(1.3,4.5)vs.3.9(2.4,5.5),P<0.05]and a lower mortality rate(11.5%vs.39.3%,P<0.05)compared to the control group.However,there were no statistically significant differences between the two groups in the incidence of postoperative intracranial hemorrhage,intracranial infection,or epilepsy(P>0.05).Conclusion Internal decompression based on white matter tract protection combined with DC can reduce mortality and contribute to improving function outcomes in patients with MMCAI.

Objective To evaluate the clinical utility of hematoma drainage supported by a multi scenario artificial intelligence(AI)system in the treatment of minor hypertensive thalamic hemorrhage(HTH)with significant neurological dysfunction.Method A retrospective analysis was conducted on minor HTH patients with significant neurological damage and incomplete interruption of corticospinal tract(CST)on the affected side.Patients in the study group received Robot of Stereotactic Surgery(ROSS)assisted drainage surgery supported by AI systems in multiple scenarios such as image reading,surgery,and follow-up.The control group received standard pharmacological treatment.Comparative analyses were performed between two groups regarding regional cerebral blood flow(rCBF),changes in rCBF(ΔrCBF),Glasgow Coma Scale(GCS),changes in GCS(ΔGCS)after 14 days of treatment and National Institutes of Health Stroke Scale(NIHSS),changes in NIHSS(ΔNIHSS),Modified Rankin Scale(mRS),changes in mRS(ΔmRS)after 90 days of treatment.to evaluate the application value of multi scenario AI system supported hematoma drainage surgery in this patient population.Results A total of 88 cases were included,with 41 in the study group and 47 in the control group.The GCS and ΔGCS values of the study group and the control group at 14 days of post-treatment were 13.4±1.6 vs 13.3±1.5 and 1.1(0.8,1.9)vs.1.1(0.9,1.8),respectively with no statistically significant difference(P>0.05).At 14 days post-treatment,statistically significant difference were observed in the rCBF of the posterior limb of the internal capsule[(379.0±55.1)mL/(kg·min)vs.(263.0±34.4)mL/(kg·min)]and the ΔrCBF[(240.0±60.6)mL/(kg·min)vs.(121.0±43.9)mL/(kg·min)].At 90 day post-treatment,statistically significant difference were also found in the NIHSS scores(3.5±1.6 vs.6.4±2.1),ΔNIHSS(-4.1±1.7 vs.-0.9±0.6),mRS scores(1.1±0.3 vs.2.3±1.0),and ΔmRS[-2.1(-2.9,-1.1)vs.-0.8(-1.4,-0.6)](P<0.05).Conclusion Hematoma drainage surgery supported by multi scenario AI systems can significantly benefit minor HTH patients with significant neurological damage and incomplete interruption of CST on the affected side.

Objective To investigate the effectiveness and safety of infarct core resection combined with decompressive craniectomy(DC)based on corticospinal tract(CST)protection in the treatment of massive cerebral infarction(MCI)with malignant brain edema.Methods This study retrospectively enrolled MCI patients with malignant brain edema who underwent internal decompression combined with DC at Xingtai Central Hospital from January 2021 to June 2024.The enrolled patients were divided into a control group and an experimental group base on the intracranial internal decompression method used.All patients underwent CT perfusion(CTP),CT angiography(CTA),diffusion-weighted imaging(DWI),and diffusion tensor imaging(DTI)within 24 h of admission.Preoperative imaging data was automatically processed using an artificial intelligence diagnostic system.For the experimental group,the imaging data was fused within a neuro-navigation system preoperatively to visualize the spatial relationships between the infarct core,ischemic penumbra,and CST and infarct core resection combined with DC was performed while protecting the CST through neuro-navigation.The control group underwent anterior temporal lobectomy combined with DC.Baseline and clinical data were collected from both groups,including gender,age,smoking history,alcohol consumption history,diabetes,hypertension,hyperlipidemia,hyperhomocysteinemia,atrial fibrillation history,responsible occluded vessel(internal carotid artery,middle cerebral artery),preoperative infarct volume on DWI,preoperative ischemic penumbra volume,preoperative the National Institutes of Health stroke scale(NIHSS)score,time from onset to surgery,intraoperative procedure duration,intraoperative blood loss,preoperative and 1-month postoperative fraction anisotropy(FA)values of the CST on the affected side,modified Rankin scale(mRS)score at 6 months postoperatively,and surgery-related complications within 1 month postoperatively(intracranial hemorrhage[operative site oozing,hemorrhagic transformation]and intracranial infection[surgical incision site infection,empyema,brain abscess,meningitis]).6-month follow-up after surgery were conducted through outpatient visit or telephone calls and prognosis of patients was evaluated using the mRS(with mRS of 0-3 defined as good prognosis,4-6 as poor prognosis,and 6 indicating death).The effectiveness indicators included FA value of the affected CST at 1 month postoperatively,good prognosis rate after surgery at 6 months,and 6-month mortality rate after surgery.The safety indicators included the incidence rates of surgical complications(intracranial hemorrhage and infection)within 1 month postoperatively.Based on preoperative DTI images,all patients were further divided into a CST-intact(infarct core did not invade CST,CST morphology intact or deformed/shifted)and a CST-damaged(infarct core invaded CST,CST disrupted or interrupted)subgroup for analysis.Results A total of 62patients(37 males,25 females,age 49-60 years,mean[55±4]years)were enrolled in this study.With 28 patients in the experimental group and 34 in the control group.(1)No significant differences were found in baseline or clinical data between the experimental and control groups(all P＞0.05),and the reoperative FA values of the affected CST were showed no significant differences(P=0.588).(2)The efficacy and safety metrics were evaluated.For the efficacy indices,at 1 month after the surgery,FA values of the affected CST increased significantly compared to preoperative values in both groups(0.409±0.051 vs.0.312±0.052 in the experimental group,and,0.381±0.048 vs.0.319±0.049 in control group;both P＜0.05),and the FA value was significantly higher in the experimental group than that in the control group(0.409±0.051 vs.0.381±0.048,P=0.030).At the 6-month follow-ups,the good prognosis rate was significantly higher in the experimental group than that in the control group(39.3％[11/28]vs.14.7％[5/34],P=0.028).No significant difference in the 6-month mortality rate were observed between the two groups(P=0.787).For the safety indices,no significant differences were found in the incidence rates of intracranial hemorrhage or intracranial infection within 1 month postoperatively between the two groups(both P＞0.05).(3)For further subgroup analysis,no significant differences were found in baseline or clinical data between the CST-damaged subgroup and the CST-intact subgroup in both the experimental and control groups(all P＞0.05).In CST-intact subgroup,FA values of the affected CST increased significantly at 1 month postoperatively compared to preoperatively in the study group(0.428±0.047 vs.0.342±0.045,P＜0.05)and the control group(0.401±0.051 vs.0.347±0.048,P＜0.05).While in the CST-damaged subgroup,no significant differences were found in FA value of the affected CST 1 month postoperatively compared with that preoperatively in both the experimental and control groups(bothP＞0.05).A significantly higher FA values 1 month postoperatively(0.428±0.047 vs.0.401±0.051,P=0.036)and good prognosis rate(9/12 vs.4/16,P=0.020)were observed in the CST-intact subgroup of the experimental group comparing with that of the control group,while there was no statistically significant difference in the 6-month mortality rate between the groups within the CST-intact subgroup(P=1.000).There were no statistically significant differences between the experimental group and the control group in both efficacy and safety indices within the CST-damaged subgroup(all P＞0.05).Conclusions Infarct core resection combining DC with CST protection demonstrates superior neurological functional improvement in comparison with anterior temporal lobectomy combining DC in treating MCI with malignant brain edema,particularly for patients with an intact CST before surgery(as indicated in patients'preoperative imaging results).This(infarct core resection combining DC with CST protection)approach does not increase the incidence of surgical complications.Prospective large sample controlled studies are required for further validation.

Objective To investigate the clinical efficacy of internal decompression based on white matter tract preservation in the treatment of malignant middle cerebral artery infarction(MMCAI).Methods A retrospective analysis was conducted on 54 patients with MMCAI.Patients were divided into a study group(n=26)and a control group(n=28)according to the surgical approach.Patients in the study group underwent preoperative fusion of CT,CTP,DWI,and DTI imaging data within a neuronavigation system.This fusion visualized the spatial relationships between the infarct core(IC),ischemic penumbra,and the corticospinal tract(CST).Subsequently,IC resection combined with decompressive craniectomy(DC)was performed while protecting the CST.Patients in the control group underwent DC alone.Key outcome measures included:changes in fractional anisotropy(FA)within the affected CST projection area at 1 month postoperatively;and 6-month postoperative mRS score,mortality,and surgical complications at 6 months postoperatively.Results At 1 month postoperatively,FA in the affected CST projection area were significantly higher in the study group than in the control group(0.092±0.013 vs.0.082±0.008,P<0.05).At the 6-month follow-up,the postoperative mRS score in the study group was significantly lower than that in the control group[2.3(1.3,4.5)vs.3.9(2.4,5.5),P<0.05]and a lower mortality rate(11.5%vs.39.3%,P<0.05)compared to the control group.However,there were no statistically significant differences between the two groups in the incidence of postoperative intracranial hemorrhage,intracranial infection,or epilepsy(P>0.05).Conclusion Internal decompression based on white matter tract protection combined with DC can reduce mortality and contribute to improving function outcomes in patients with MMCAI.

Objective To evaluate the application value of stereotactic robotic surgery driven by a multi scenario AI system in the elderly with severe hypoperfusion in the internal capsule area and minor HTH.Methods A retrospective analysis was conducted on 101 minor HTH patients com-plicated with severe hypoperfusion in the internal capsule area identified by an AI diagnostic sys-tem in our department from January to October 2024.Among them,48 cases who underwent ster-eotactic robotic drainage with AI system support were assigned into the study group,and 53 cases of conventional drug treatment were into the control group.The general clinical data were com-pared between the two group.Improvement rate of rCBF 14 d after treatment,effective rate 90 d after treatment,favorable prognosis rate,and incidence of shunt dependent hydrocephalus were compared between the two groups to evaluate the application value of AI-assisted stereotactic ro-botic surgery in these patients.Results The study group had significantly higher improvement rate of rCBF 14 d after treatment,better effective rate and larger ratio of favorable prognosis 90 d after treatment when compared with the control group(60.4％vs 39.6％,P＜0.05;62.5％vs 41.5％,P＜0.05;64.6％vs 43.4％,P＜0.05).Conclusion Stereotactic robotic assisted drainage driven by multi scenario AI system can significantly improve the prognosis of the elderly minor HTH patients complicated with severe hypoperfusion in the internal capsule area.

Objective To evaluate the application value of stereotactic robotic surgery driven by a multi scenario AI system in the elderly with severe hypoperfusion in the internal capsule area and minor HTH.Methods A retrospective analysis was conducted on 101 minor HTH patients com-plicated with severe hypoperfusion in the internal capsule area identified by an AI diagnostic sys-tem in our department from January to October 2024.Among them,48 cases who underwent ster-eotactic robotic drainage with AI system support were assigned into the study group,and 53 cases of conventional drug treatment were into the control group.The general clinical data were com-pared between the two group.Improvement rate of rCBF 14 d after treatment,effective rate 90 d after treatment,favorable prognosis rate,and incidence of shunt dependent hydrocephalus were compared between the two groups to evaluate the application value of AI-assisted stereotactic ro-botic surgery in these patients.Results The study group had significantly higher improvement rate of rCBF 14 d after treatment,better effective rate and larger ratio of favorable prognosis 90 d after treatment when compared with the control group(60.4％vs 39.6％,P＜0.05;62.5％vs 41.5％,P＜0.05;64.6％vs 43.4％,P＜0.05).Conclusion Stereotactic robotic assisted drainage driven by multi scenario AI system can significantly improve the prognosis of the elderly minor HTH patients complicated with severe hypoperfusion in the internal capsule area.

Objective To evaluate the clinical utility of hematoma drainage supported by a multi scenario artificial intelligence(AI)system in the treatment of minor hypertensive thalamic hemorrhage(HTH)with significant neurological dysfunction.Method A retrospective analysis was conducted on minor HTH patients with significant neurological damage and incomplete interruption of corticospinal tract(CST)on the affected side.Patients in the study group received Robot of Stereotactic Surgery(ROSS)assisted drainage surgery supported by AI systems in multiple scenarios such as image reading,surgery,and follow-up.The control group received standard pharmacological treatment.Comparative analyses were performed between two groups regarding regional cerebral blood flow(rCBF),changes in rCBF(ΔrCBF),Glasgow Coma Scale(GCS),changes in GCS(ΔGCS)after 14 days of treatment and National Institutes of Health Stroke Scale(NIHSS),changes in NIHSS(ΔNIHSS),Modified Rankin Scale(mRS),changes in mRS(ΔmRS)after 90 days of treatment.to evaluate the application value of multi scenario AI system supported hematoma drainage surgery in this patient population.Results A total of 88 cases were included,with 41 in the study group and 47 in the control group.The GCS and ΔGCS values of the study group and the control group at 14 days of post-treatment were 13.4±1.6 vs 13.3±1.5 and 1.1(0.8,1.9)vs.1.1(0.9,1.8),respectively with no statistically significant difference(P>0.05).At 14 days post-treatment,statistically significant difference were observed in the rCBF of the posterior limb of the internal capsule[(379.0±55.1)mL/(kg·min)vs.(263.0±34.4)mL/(kg·min)]and the ΔrCBF[(240.0±60.6)mL/(kg·min)vs.(121.0±43.9)mL/(kg·min)].At 90 day post-treatment,statistically significant difference were also found in the NIHSS scores(3.5±1.6 vs.6.4±2.1),ΔNIHSS(-4.1±1.7 vs.-0.9±0.6),mRS scores(1.1±0.3 vs.2.3±1.0),and ΔmRS[-2.1(-2.9,-1.1)vs.-0.8(-1.4,-0.6)](P<0.05).Conclusion Hematoma drainage surgery supported by multi scenario AI systems can significantly benefit minor HTH patients with significant neurological damage and incomplete interruption of CST on the affected side.

Objective:To evaluate the application value of artificial intelligence (AI) and robot of stereotactic surgery (ROSS)-assisted ultra-early thalamic hematoma drainage in elderly patients with minor thalamus and internal capsule hematoma (TICH) and severe hypoperfusion in the internal capsule area.Methods:A retrospective cross-sectional study was performed; 278 patients with TICH and severe hypoperfusion in the internal capsule area identified by AI medical imaging diagnostic system were enrolled from Department of Neurosurgery, Xingtai Central Hospital from January 2023 to August 2024. Among them, 134 patients (study group) received AI and ROSS-assisted ultra-early thalamic hematoma drainage, and 144 patients (control group) received drug treatment. Differences in baseline data and therapy efficiency between the two groups of patients were compared. Multivariate Logistic regression analysis was used to screen the independent influencing factors for prognosis in the study group 180 days after onset (good prognosis: modified Rankin scale scores of 0-3).Results:Compared with the control group, the study group had significantly higher Glasgow coma scale (GCS) score and regional cerebral blood flow (rCBF) in the internal capsule area 14 days after onset, and statistically higher therapy efficiency and good prognosis rate 180 days after onset ( P<0.05). In the study group, 84 patients had a good prognosis and 50 had a poor prognosis 180 days after onset; compared with the good prognosis group, the poor prognosis group had significantly higher proportion of patients with hyper-homocysteinemia history, lower GCS score on admission, larger volume of thalamic hematoma, higher proportions of patients with combined intraventricular hemorrhage and shunt-dependent hydrocephalus, and lower rCBF in the internal capsule area on admission ( P<0.05). Multivariate Logistic regression analysis showed that thalamic hematoma volume ( OR=2.527, 95% CI: 1.504-4.247, P<0.001), combined intraventricular hemorrhage ( OR=2.325, 95% CI: 1.460-3.703, P<0.001), shunt dependent hydrocephalus ( OR=2.371, 95% CI: 1.267-4.078, P=0.006), and rCBF in the internal capsule area on admission ( OR=0.672, 95% CI: 0.314-1.025, P<0.001) were independent risk factors for prognosis of patients in the study group. Conclusion:AI and ROSS-assisted ultra-early thalamic hematoma drainage is effective for elderly patients with minor TICH and severe hypoperfusion in the internal capsule area; patients with large thalamic hematoma volume, combined intraventricular hemorrhage, shunt-dependent hydrocephalus, and low rCBF in the internal capsule area on admission are prone to have a poor prognosis.